Autumn 2013

Acidizing Advances

Monitoring Casing Corrosion

Geomagnetic Referencing

Solar Storms

Oilfield Review

13-OR-0004

Oilfield Review AppSchlumberger Oilfield Review iPad† app for the Newsstand is available free of charge at the Apple† iTunes† App Store.

Oilfield Review communicates advances in finding and producing hydrocarbons to oilfield professionals. The free Oilfield Review Apple iPad app for accessing content is part of the Newsstand and allows access to both new and archived issues. Many articles have been augmented with richer content such as animations and videos, which help explain concepts and theories beyond the capabilities of static images. The app offers access to several years of archived issues in a compact format that retains the high-quality images and content you’ve come to expect from the print version of Oilfield Review.

Download and install the app from the iTunes App Store by searching for “Schlumberger Oilfield Review” from your iPad or scan the QR code below, which will take you directly to the iTunes site.

†Apple, iPad, and iTunes are marks of Apple Inc., registered in the U.S. and other countries.

Announcing the new

Oilfield Review iP

ad app

Widely hailed as a breakthrough technology, extended-reach drilling enables cost-effective development of off-shore reserves from shore-based locations and centrally located platforms. In addition, it achieves maximum con-tact with the reservoir and accesses multiple reservoirs with a single wellbore.

For Eni US Operating Company Inc., extended-reach drilling has been instrumental in our development of the Nikaitchuq field off the North Slope of Alaska, USA. The field’s characteristics—from its offshore location and downhole temperature to its complex geology—make this a highly challenging project.

Our leases are offshore in the Arctic Ocean and the Beaufort Sea north of the Arctic Circle. For better access to the reservoir, we’ve built an island a few miles offshore, in less than 10 ft [3 m] of water. Of the 30 wells we plan to drill from the single pad on the man-made island, we have drilled 17; we have also drilled 22 from an onshore pad located at Oliktok Point. In addition to the economic bene-fits, restricting drilling sites to just two pads minimizes our environmental footprint.

The reservoir we are targeting is shallow and relatively cool, making the oil viscous. This reservoir had been devel-oped as a line-drive waterflood for optimal oil recovery; the development plan features alternating horizontal injectors and producers, with a total of 52 wells to be completed by 2014. The injection water for waterflood is produced from a deeper, warmer formation.

We are drilling shallow, extended-reach wells. While these wells are between 3,200 ft [1,000 m] and 4,200 ft [1,300 m] deep, some are more than 23,000 ft [7,000 m] long. More than 90% of the wells in the development have a reach/TVD ratio of more than 4 and some are as high as 6. The wells are spaced 1,200 ft [370 m] apart along their production intervals, and several follow faults that compartmentalize the reservoir. Accurate well placement is crucial to ensure we don’t short-circuit the waterflood or inadvertently cross a major fault. A 1% location error in a 23,000-ft long well translates into an unacceptable error of more than 200 ft [60 m] at TD.

This is where geomagnetic referencing comes in. Although traditional gyroscopic surveys could produce data of sufficient quality to achieve the necessary well-bore placement positions, gyro surveys are impractical in this environment and require additional costs and time that make them prohibitively expensive for drilling pro-grams in this area. Geomagnetic referencing provides us with real-time, precise positioning and the certainty of knowing where our wellbores are without having to stop

Geomagnetic Referencing for Well Placement

1

the drilling process. By using geomagnetic referencing, we are able to construct a detailed model of the Earth’s magnetic field for comparison with magnetic measure-ments acquired while drilling (see “Geomagnetic Referencing—The Real-Time Compass for Directional Drillers,” page 32). The model is made up of contribu-tions from the Earth’s main magnetic field, the local mag-netic variations in crustal rocks and time-varying disturbances caused by solar activity.

Solar-related magnetic storms occur unpredictably, and at Arctic latitudes, they generate high-amplitude swings in magnetic field strength and direction that must be incor-porated into the model. To quantify these disturbances, Schlumberger partnered with the US Geological Survey to build a geomagnetic observatory nearby in Deadhorse, Alaska. The observatory supplies the high-quality referenc-ing data required for real-time drill-ahead corrections and for definitive surveys at the end of each bottomhole assem-bly run.

We are drilling our 39th well using geomagnetic refer-encing. Since the earliest applications of this technology in our wells, our wellbore position uncertainty has continually decreased. And because we know the positions with a high degree of certainty, we are reentering wells to create dual laterals from single laterals. This strategy allows us to essentially double the wellbore contact with the reservoir and increase production rates. Even with these increased rates, we expect to produce from this field for more than 30 years.

Andrew BuchananSenior Operations GeologistEni US Operating Company Inc.Anchorage, Alaska, USA

Andrew Buchanan is the Senior Operations Geologist with Eni US Operating Company Inc. in Anchorage, where he has been since 2009. He previously worked for ASRC Energy Services as a geologic consultant. Andrew earned a BS degree in geology from the University of Regina, Saskatchewan, Canada. He currently serves as Past President of the Petroleum Club of Anchorage.

www.slb.com/oilfieldreview

Schlumberger

Oilfield Review

1 Geomagnetic Referencing for Well Placement

Editorial contributed by Andrew Buchanan, Senior Operations Geologist, Eni US Operating Company Inc.

4 Stimulating Naturally Fractured Carbonate Reservoirs

Stimulation of naturally fractured carbonate reservoirs has improved significantly with the application of innovative acidizing fluids that contain degradable fibers. The fibers congregate and form barriers that impede fluid movement into fractures, redirecting the acid to lower permeability regions. This type of enhanced stimulation efficiency has led to increasingly uniform production profiles across multiple zones and substantial production increases in many oil and gas fields worldwide.

18 Casing Corrosion Measurement to Extend Asset Life

Corrosion in downhole tubulars may shorten a well’s produc-tive life and contribute to costly damages for operators. Downhole corrosion monitoring serves as the first line of defense against casing corrosion.

Executive EditorLisa Stewart

Senior EditorsTony SmithsonMatt VarhaugRick von Flatern

EditorRichard Nolen-Hoeksema

Contributing EditorsH. David LeslieTed MoonParijat MukerjiErik NelsonGinger OppenheimerRana Rottenberg

Design/ProductionHerring DesignMike Messinger

Illustration Chris LockwoodMike MessingerGeorge Stewart

PrintingRR Donnelley—Wetmore PlantCurtis Weeks

Oilfield Review is published quarterly and printed in the USA.

Visit www.slb.com/oilfieldreview for electronic copies of articles in English, Spanish, Chinese and Russian. A free iPad® app is available for download.

© 2013 Schlumberger. All rights reserved. Reproductions without permission are strictly prohibited.

For a comprehensive dictionary of oilfield terms, see the Schlumberger Oilfield Glossary at www.glossary.oilfield.slb.com.

About Oilfield ReviewOilfield Review, a Schlumberger journal, communicates technical advances in finding and producing hydrocarbons to customers, employees and other oilfield professionals. Contributors to articles include industry professionals and experts from around the world; those listed with only geographic location are employees of Schlumberger or its affiliates.

On the cover:

The aurora borealis appears as shim-mering curtains of colored light in the Arctic regions of the Earth’s northern hemisphere. Auroras, which may occur in both of the Earth’s polar regions, are created when emissions from solar flares and coronal mass ejections inter-act with the Earth’s magnetic field. A large loop of plasma, referred to as a prominence, emanates from the Sun’s surface (inset). Such a mass of plasma ejected in the direction of the Earth would create space weather events that could disrupt modern electromag-netic-related technologies, including well guidance methods that depend on magnetic measurements.

2

Autumn 2013Volume 25Number 3

ISSN 0923-1730

61 Contributors

63 Defining Production Logging: Principles of Production Logging

This is the eleventh in a series of introductory articles describing basic concepts of the E&P industry.

3

32 Geomagnetic Referencing—The Real-Time Compass for Directional Drillers

In recent years, demand for accurate wellbore placement has driven technology developments that have advanced the science of wellbore guidance. This article examines magnetic surveying methods that improve real-time measurement accuracy and allow drillers to reach their targets efficiently and cost-effectively.

48 Blowing in the Solar Wind: Sun Spots, Solar Cycles and Life on Earth

Space weather can affect terrestrial systems that are crucial for modern society. This article describes solar events that contribute to space weather and are the source of electro-magnetic pulses that have the potential to disrupt and damage electronic, power, communication, transportation and other infrastructure technologies on Earth and in space. Solar sunspot cycles and their influence on solar and terres-trial weather are also discussed.

Hani Elshahawi Shell Exploration and Production Houston, Texas, USA

Gretchen M. Gillis Aramco Services Company Houston, Texas

Roland Hamp Woodside Energy Ltd. Perth, Australia

Dilip M. Kale ONGC Energy Centre Delhi, India

George King Apache Corporation Houston, Texas

Andrew Lodge Premier Oil plc London, England

Advisory Panel

Editorial correspondenceOilfield Review 5599 San FelipeHouston, TX 77056United States(1) 713-513-1194Fax: (1) 713-513-2057E-mail: editorOilfieldReview@slb.com

SubscriptionsCustomer subscriptions can be obtained through any Schlumberger sales office. Paid subscriptions are available fromOilfield Review ServicesPear Tree Cottage, Kelsall RoadAshton Hayes, Chester CH3 8BHUnited KingdomE-mail: subscriptions@oilfieldreview.com

Distribution inquiriesMatt VarhaugOilfield Review 5599 San FelipeHouston, TX 77056United States(1) 713-513-2634E-mail: DistributionOR@slb.com

4 Oilfield Review

Stimulating Naturally Fractured Carbonate Reservoirs

Naturally fractured carbonate reservoirs can be difficult to stimulate because

treatment fluids tend to enter the fractures and avoid less permeable regions.

Effective fluid diversion techniques are usually necessary to ensure that stimulation

fluids contact the largest possible reservoir surface area. Engineers and chemists

have developed an innovative acidizing fluid that employs degradable fibers to

temporarily block permeable fractures and force the fluid into less permeable zones.

Operators have applied the fiber-laden acid to naturally fractured oil and gas reser-

voirs in which achieving complete zonal coverage is difficult and, as a result, have

witnessed substantial production improvements.

Khalid S. AsiriMohammed A. AtwiSaudi AramcoUdhailiyah, Saudi Arabia

Oscar Jiménez BuenoPetróleos Mexicanos (PEMEX)Villahermosa, Mexico

Bruno LecerfAlejandro PeñaSugar Land, Texas, USA

Tim LeskoConway, Arkansas, USA

Fred MuellerCollege Station, Texas

Alexandre Z. I. PereiraPetrobrasRio de Janeiro, Brazil

Fernanda Tellez CisnerosVillahermosa, Mexico

Oilfield Review Autumn 2013: 25, no. 3. Copyright © 2013 Schlumberger.For help in preparation of this article, thanks to Charles-Edouard Cohen, Rio de Janeiro; Alan Diaz and Victor Ariel Exler, Macaé, Brazil; Luis Daniel Gigena, Mexico City; Daniel Kalinin, Al-Khobar, Saudi Arabia; and Svetlana Pavlova, Novosibirsk, Russia.ACTive, MaxCO3 Acid, POD, SXE and VDA are marks of Schlumberger.

1. Crowe C, Masmonteil J, Touboul E and Thomas R: “Trends in Matrix Acidizing,” Oilfield Review 4, no. 4 (October 1992): 24–40.

2. Robert JA and Rossen WR: “Fluid Placement and Pumping Strategy,” in Economides MJ and Nolte KG (eds): Reservoir Stimulation, 3rd ed. Chichester, West Sussex, England: John Wiley & Sons, Ltd (2000): 19-2–19-3.

Autumn 2013 55

Since the dawn of the oil and gas industry, opera-tors have endeavored to maximize well productiv-ity, employing a variety of techniques to do so. For example, as early as the 19th century, engineers began pumping acid in wells to improve produc-tion. Acidizing treatments dissolve and remove formation damage resulting from drilling and completion operations, create new production pathways in producing formations or both.

Acidizing treatments fall into two categories. Matrix acidizing consists of pumping fluid into the formation at rates and pressures that will not fracture the reservoir. The resulting treatment stimulates a region extending up to about 1 m [3 ft] around the wellbore. Fracture acidizing is a hydraulic fracturing treatment that pumps acid during at least one fluid stage. The stimulation distance may extend one or two orders of magni-tude farther into the formation than that achieved by matrix acidizing.

The composition of acidizing fluids depends on the type of formation to be stimulated. Carbonate formations, composed mainly of lime-stone (calcium carbonate [CaCO3]) or dolomite (calcium magnesium carbonate [CaMg(CO3)2]), are treated with hydrochloric acid [HCl], various organic acids or combinations thereof. Sandstone formations typically consist of quartz [SiO2] or feldspar [KAlSi3O8–NaAlSi3O8–CaAl2Si2O6] par-ticles bound together by carbonate or clay miner-als. Silicate minerals do not react with HCl; they respond instead to stimulation fluids that contain hydrofluoric acid [HF] or fluoboric acid [HBF4].1 Despite the fluid chemistry differences, the engi-neering aspects of carbonate and sandstone acidizing are largely similar. However, this article concentrates on recent advances that are partic-ularly relevant to carbonate acidizing.

Carbonate Acidizing FundamentalsLimestone and dolomite rapidly dissolve in HCl, forming water-soluble reaction products—mainly calcium and magnesium chlorides—and liberating carbon dioxide. The dissolution rate is limited by the speed at which acid can be delivered to the rock surface. This dissolution process results in rapid formation of irregularly shaped channels called wormholes (above right). Wormholes radiate outward in a dendritic pat-tern from points where acid leaves the well and enters the formation. Once formed, they become the most permeable pathways into the formation and carry virtually all of the fluid flow during pro-duction. For efficient stimulation, the wormhole network should penetrate deeply and uniformly throughout the producing interval.

Achieving stimulation uniformity can be par-ticularly challenging when large permeability variations exist within the treatment interval. As acid penetrates the formation, it flows preferen-tially into the most-permeable pathways. Higher-permeability areas receive most of the fluid and become larger, causing the treatment fluids to bypass lower-permeability regions where stimu-lation is needed most. To address this problem, engineers and chemists have developed methods

to divert acidizing fluids away from high-permea-bility intervals and into less permeable zones.

Engineers accomplish diversion by employing mechanical or chemical means or both.2 Mechanical diversion of treatment fluids may be achieved using drillpipe or coiled tubing–con-veyed tools equipped with mechanical packers that isolate and direct fluid into low-permeability zones. Alternatively, flow can be blocked at indi-vidual perforations by dropping ball sealers into

> Acid-induced wormholes. An intricate network of wormholes formed during a laboratory-scale matrix acidizing treatment of a carbonate formation sample. The length, direction and number of wormholes depend on the formation reactivity and the rate at which acid enters the formation. Once formed, the wormholes may carry virtually all of the fluid flow during production.

6 Oilfield Review

the stimulation fluid as it travels down the well. The ball sealers are drawn to and seat against perforations accepting the most fluid. After the treatment, the ball sealers fall away, are mechan-ically dislodged or dissolve (above).

Chemical diverting agents incorporated in stimulation fluids may be divided into two catego-ries—particulates and viscosifiers. Particulates include plugging agents such as benzoic acid flakes and salt grains that are sized to plug forma-tion pores. Foaming the acid may achieve a simi-lar plugging effect because of two-phase flow.

Viscosifiers include water-soluble polymers, crosslinked polymer gels and viscoelastic surfac-tants (VESs).3 A decade ago, Schlumberger scien-tists and engineers applied VES chemistry to acid stimulation and introduced the VDA viscoelastic

diverting acid system. VDA fluids have been par-ticularly successful in both matrix and fracture acidizing applications around the world.4

The surfactant molecule in the VDA system, derived from a long-chain fatty acid, is zwitter-ionic—a neutral molecule that carries a positive and a negative charge at separate positions.5 While being pumped down a well, VDA fluid—a blend of HCl, VES and common acid-treatment additives—maintains a low viscosity. As the acid is consumed in the formation, the surfactant mol-ecules begin to aggregate into elongated micelles.6 The micelles become entangled and cause the fluid viscosity to increase (below). The higher-viscosity fluid forms a temporary barrier that forces fresh acid to flow elsewhere. In addi-tion to providing diversion, the viscosity decreases

the rate at which the acid reacts with the forma-tion, thereby allowing more time for the creation of deeper and more intricate wormholes.

When production begins, VDA fluid is exposed to hydrocarbons, which alters the ionic environ-ment and causes the micelles to become spheri-cal. Entanglement ceases, the micelles roam freely, and the fluid viscosity decreases dramati-cally, enabling efficient poststimulation cleanup. Unlike polymer-base fluids, VESs leave virtually no damaging residue behind that may interfere with well productivity.

Naturally fractured reservoirs are the most challenging environments for carbonate acidiz-ing because they can present extreme permeabil-ity contrasts. The fractured regions may be several orders of magnitude more permeable than the unfractured layers. Until recently, the industry’s considerable portfolio of diversion technologies has been inefficient in this environ-ment. Even when using self-diverting fluids such as the VDA formulation, engineers struggled to block the fractures and treat the rest of the for-mation. Consequently, operators were forced to pump large volumes of fluid to achieve stimula-tion, leading to higher treatment costs and less than optimal results.

However, Schlumberger engineers and chem-ists discovered that significant diversion improve-ments could be achieved by adding degradable fibers to VDA fluid. As fiber-laden diversion fluid enters a fracture, the fibers congregate, entangle and form structures that limit fluid entry. The new product, MaxCO3 Acid degradable diversion acid system, has been used successfully and effi-ciently to stimulate notoriously difficult carbon-ate reservoirs around the world.

>Mechanical diversion methods. Ball sealers (green spheres) are pumped down the well during the stimulation treatment (left). The balls provide mechanical diversion because they preferentially block the perforations that take the highest volume of treatment fluid. Straddle packers may also be deployed on coiled tubing to isolate the preferred treatment interval (right). In this example, engineers have already stimulated the bottom zone and moved the packers up in preparation for stimulating the next zone.

Ball Sealers Straddle Packers

> Viscoelastic surfactant (VES) fluid behavior during an acidizing treatment. Initially, when the surfactant is dispersed in acid, each molecule moves independently throughout the fluid (left). As the acid reacts with the carbonate minerals, the surfactant molecules assemble and create elongated micelles (center). The micelles entangle and hinder fluid flow, resulting in higher fluid viscosity. When hydrocarbon production begins after the treatment, the elongated micelles transform into spheres (right), resulting in a dramatic decrease in fluid viscosity and facilitating efficient cleanup.

Surfactantmolecules

Elongated micelles Spherical micelles

Spent acid Hydrocarbon

CaCO3 + 2HCl CaCl2 + CO2 + H2O

Autumn 2013 7

This article describes the development of the MaxCO3 Acid system in the laboratory and its introduction to the oil field. Case histories from Mexico, Saudi Arabia and Brazil demonstrate how application of this new acid system is achiev-ing significant well productivity improvements.

Studying Fiber-Laden Acids in the LaboratoryFor more than 20 years, chemists and engineers have explored ways in which fibers could be used to improve well servicing operations. Working

with both mineral- and polymer-base fibers, they discovered techniques for controlling the behav-ior of fluids and suspended solids, both during and after placement in a well. The research resulted in several innovations, including meth-ods for limiting lost circulation during drilling and cementing, improving the flexibility and durability of well cements, aiding proppant trans-port during hydraulic fracturing operations and preventing proppant flowback into the well after a fracturing treatment.

Studying applications for fibers in the context of acidizing has been a more recent endeavor. In 2007, scientists at Schlumberger began exploring the ability of fibers to improve fluid diversion in both openhole and cased hole scenarios (above). The principal difference between the two condi-tions is that, for openhole completions, fibers must accumulate along the entire wellbore sur-face to provide diversion, but in a cased hole situation, fiber deposition may be confined to perforations.

The engineers discovered that simply adding fibers to a conventional HCl solution failed to cre-ate a stable fibrous suspension. Shortly after addition, the fibers congregated, formed clumps and separated from the acid. Success was achieved by adding fibers to VDA fluid. The resul-tant higher fluid viscosity allowed the creation of a robust suspension of discrete fibers.

3. For more on water-soluble polymers and VESs: Gulbis J and Hodge RM: “Fracturing Fluid Chemistry and Proppants,” in Economides MJ and Nolte KG (eds): Reservoir Stimulation, 3rd ed. Chichester, West Sussex, England: John Wiley & Sons, Ltd (2000): 7-1–7-23.

4. Al-Anzi E, Al-Mutawa M, Al-Habib N, Al-Mumen A, Nasr-El-Din H, Alvarado O, Brady M, Davies S, Fredd C, Fu D, Lungwitz B, Chang F, Huidobro E, Jemmali M, Samuel M and Sandhu D: “Positive Reactions in Carbonate Reservoir Stimulation,” Oilfield Review 15, no. 4 (Winter 2003/2004): 28–45.

Lungwitz B, Fredd C, Brady M, Miller M, Ali S and Hughes K: “Diversion and Cleanup Studies of Viscoelastic

> Fiber deposition and diversion scenarios. During openhole acidizing (top and bottom left), fibers form a filtercake that covers the entire wellbore wall. During cased hole acidizing (top and bottom right), fibers form filtercakes in the perforation tunnels.

Wellborewall

Openhole Acidizing Cased Hole Acidizing

Well Well

Casing

Filtercake

Filtercake

FiltercakeTreatment fluid Treatment fluid

Filtercake

Wormhole

Wormhole Perforation

Perforation

Casing

Surfactant-Based Self-Diverting Acid,” SPE Production & Operations 22, no. 1 (February 2007): 121–127.

5. Sullivan P, Nelson EB, Anderson V and Hughes T: “Oilfield Applications of Giant Micelles,” in Zana R and Kaler EW (eds): Giant Micelles—Properties and Applications. Boca Raton, Florida, USA: CRC Press (2007): 453–472.

6. A micelle is a colloidal assembly of surfactant molecules. In the aqueous environment of an acidizing fluid, the surfactant molecules are arranged such that the interior of the micelle is hydrophobic and the exterior is hydrophilic. Worm-like micelles may be microns long and have a cross section of a few nanometers.

8 Oilfield Review

The engineers then began performing exper-iments with laboratory-scale equipment for simulating fluid leakoff and fiber deposition (above). The principal simulator was a bridging apparatus that accommodated a variety of ori-fices through which fiber-laden acid could pass at various flow rates. Circular orifices, with diameters between 1 and 2 mm [0.04 and 0.08 in.], simulated wormholes. Rectangular ori-fices with widths between 2 and 6 mm [0.08 and 0.24 in.] were analogous to fractures. Engineers observed fiber plug formation and recorded the corresponding system pressure as fiber-laden acid passed through an orifice.

> Laboratory-scale equipment for testing leakoff behavior and filtercake deposition. Engineers used a conventional filtration cell to simulate an openhole stimulation (top). Technicians first placed a carbonate core at the bottom of the cell and then poured in fiber-laden acid. After sealing the cell, they applied differential pressure across the core and used a balance to measure the amount of filtrate passing though the core. For the cased hole simulation (bottom), engineers used a bridging apparatus. The apparatus consisted mainly of a 300-mL tube fitted with a piston, a high-performance liquid chromatography (HPLC) pump and an orifice (left). The orifice could be circular to simulate a wormhole (top right) or rectangular to mimic a fracture (bottom right). Technicians installed a piston at the top of the tube, which contained fiber-laden acid. Acid exiting the tube passed through the orifice, and the technicians assessed the diversion capability of fibers by measuring the filtrate volume, the fiber filtercake volume and the pumping pressure at various flow rates.

Pressure

Filtercake

Filtrate

Balance

Pressure cell

Acid andfibers

Backpressureregulator

Core

Openhole Simulation

Flui

d flo

w

130 mm

ID 21 mm

20 mm1 to 2 mm

2 to 6 mm

25.75 mm

65 mm

75 mm

Piston

FiltercakeOrifice

Orifice

Orifice

Pressure sensor14

2 cm

Pump

Wormhole Geometry

Fissure or Fracture Geometry

Acidand fibers

Cased Hole Simulation

Pressure evolution in the apparatus followed a consistent pattern (next page, top left). Initially there was no pressure increase, but within a few seconds, the pressure rose rapidly as the fibers formed a bridge and began to fill the orifice. These results indicated that as early volumes of fiber-laden acid reach the perfora-tions, the acid penetrates the reservoir as if no fibers are present. Then, as the fibers bridge, they accumulate inside the perforations and form a filtercake. Next, the fibers plug the perforation, decreasing injectivity and promot-ing fluid diversion into other perforations. The engineers also discovered that the fiber

concentration required to achieve bridging increased with the fluid injection rate (next page, top right).

In the laboratory, after pumping the fiber-laden acid through the orifice, engineers per-formed a freshwater flush. As the viscous acid left the apparatus, the pumping pressure gradu-ally decreased and eventually stabilized. At the end of each test, a stable fiber plug remained in the orifice. Knowing the pressure, flow rate, fluid viscosity and fiber plug length, engineers were also able to use Darcy’s law to calculate the fiber plug permeabilities. Depending on the fiber concentration and the fluid flow rate dur-

Autumn 2013 9

7. It may appear counterintuitive to imagine that fiber plugs with permeabilities higher than that of the formation could provide significant diversion. However, significant diversion is also provided by the flow restriction and pressure drop as fluid enters the perforations.

8. Cohen CE, Tardy PMJ, Lesko T, Lecerf B, Pavlova S, Voropaev S and Mchaweh A: “Understanding Diversion with a Novel Fiber-Laden Acid System for Matrix Acidizing of Carbonate Formations,” paper SPE 134495, presented at the SPE Annual Technical Conference and Exhibition, Florence, Italy, September 19–22, 2010.

> Pressure-versus-time plot from a slot-flow experiment. During this experiment, the MaxCO3 Acid composition consisted of 15 wt% VDA fluid and 6 kg/m3 (50 lbm/1,000 galUS) degradable fibers. In Period 0, MaxCO3 Acid fluid begins flowing through the slot, and the fibers have not yet formed a bridge. In Period 1, the pressure rises as the fibers entangle and form a plug in the slot. Pressure continues to climb until the volume of acid is exhausted. In Period 2, the pressure gradually falls as freshwater enters the slot and displaces the viscous acid. The system pressure stabilizes during Period 3. The white fiber plug remains intact and stable inside the slot (photograph).

Pres

sure

, psi

40

50

60

30

0 1 2 3

20

10

0 10 20 30

Time, s40 50 60 70 80

0

2-mm slot

Fluid inflow

ing fiber deposition, the measured permeabili-ties varied between 400 and 2,400 mD. These data led engineers to conclude that fibers would provide the most efficient diversion in zones with permeabilities exceeding 100 mD (left).7

The data acquired during the simulator exper-iments also allowed scientists to develop a math-ematical model for predicting the behavior of fiber-laden acids under openhole and cased hole conditions; the model may be used to optimize treatment designs.8 They performed 340 fine-scale 3D simulations that evaluated typical perforation schemes, fibrous filtercake permeabilities and formation permeabilities. The resulting model allows scientists to track the movement of the flu-ids and fibers through the wellbore and into the reservoir and track the propagation of wormholes generated as the acid reacts with carbonate rock.

> Effect of degradable fiber concentration on bridging ability in a slot. During the slot-flow experiments, engineers determined that the fiber concentration required to achieve bridging and promote fluid diversion increases with the fluid injection rate.

Linear fluid velocity, m/min

Linear fluid velocity, ft/min

30251550 2010

32.8 49.2 65.6 82.0 98.416.40

50

100

150

Degr

adab

le fi

ber c

once

ntra

tion,

lbm

/1,0

00 g

alUS

Bridging region

Nonbridging region

> Apparent permeability resulting from plugging a perforated zone with fibers. The x-axis shows the original core permeability. The y-axis shows the apparent zone permeability after a fibrous filtercake with a permeability of 2 D has formed. The results show that after plugging occurs, when core permeability exceeds about 1 mD, apparent permeability eventually levels off at about 100 mD and becomes independent of core permeability.

Appa

rent

per

mea

bilit

y, m

D

0.10.1

1

1

10

10

100

100

10,000

10,000

1,000

1,000

Core permeability, mD

10 Oilfield Review

> Diversion predictions from the MaxCO3 Acid simulator. During fiber deposition experiments in the perforation simulator, the permeabilities of the resulting fiber plugs varied between about 400 and 2,400 mD (left). The simulator predicts how the fiber plugs decrease the apparent permeabilities of reservoirs and promote diversion. Lower-permeability fiber plugs are more efficient diverters. Modeling studies also demonstrated that fibrous filtercakes provide fluid diversion by equalizing the permeabilities of layers in the treated interval. For example, if the interval contains four layers with various permeabilities, the fluid flow rate into the more permeable layers decreases and the fluid flow rate into the less permeable layers increases. Eventually, the flow rates converge to a single flow rate, and the interval behaves as if it has a single permeability (right). Flow rate convergence occurs more quickly in a cased hole with perforations because the filtercake surface area is lower.

Appa

rent

rese

rvoi

r per

mea

bilit

y, m

D

Reservoir permeability, mD0.1

0.11

1

10

10

100

100

10,000

10,000

1,000

1,000

Fiber plug permeability2,400 mD1,500 mD400 mD

Layer permeability30 D10 D3 D1 D

Time

Flow

rate

>MaxCO3 Acid fluid batch mixing. The degradable fibers (top left) are light and finely divided, presenting a mixing challenge. Traditional equipment for batch mixing of acidizing fluids was inefficient. Engineers discovered that equipment for batch mixing cement slurries (bottom left) could disperse the fibers in VDA fluid. The VDA fluid flows into an 8,000-L [50-bbl] paddle mixer (top right). To avoid the formation of clumps, field personnel manually add fibers to the fluid. After the fibers have been added, the tank is filled with more VDA fluid, and agitation continues until the mixture reaches a uniform consistency (bottom right). During the job, engineers maintain the agitation to preserve fluid uniformity.

Autumn 2013 11

9. For more on formation damage testing in the laboratory: Hill DG, Liétard OM, Piot BM and King GE: “Formation Damage: Origin, Diagnosis and Treatment Strategy,” in Economides MJ and Nolte KE (eds): Reservoir Stimulation, 3rd ed. Chichester, West Sussex, England: John Wiley & Sons, Ltd (2000): 14-31–14-33.

> Behavior of degradable fibers. Engineers performed static bottle tests during which degradable fibers were immersed in partially spent HCl fluids. The data show that the rate of fiber dissolution decreases as the HCl becomes neutralized. Nevertheless, complete fiber dissolution occurs within a few days (top). Core testing demonstrated that the acidic fiber degradation products may further stimulate the formation (bottom). Using a standard core testing apparatus at 115°C [239°F], engineers pumped 2% KCl solution into a limestone core first in the injection direction and then in the reverse, or production, direction (K0 and K1). Technicians recorded the pressure across the core and, applying Darcy’s law, determined that the initial core permeability was 5.1 mD. Next, they injected a partially spent 20% HCl fluid (pH = 6.5) containing degradable fibers (N2). Subsequent pumping of 2% KCl in both directions revealed that the core permeability had fallen to 3.5 mD (K2 and K3). Following a 16-h shut-in period, the fibers had begun to degrade, and the core permeability rose to about 4.8 mD (K4 and K5). After another 16-h shut-in period, complete fiber degradation had occurred, and the core permeability rose to 5.5 mD (K6 and K7)—an 8% improvement over the initial permeability of 5.1 mD.

Fibe

r deg

rada

tion

time,

h

Volume of acid spent at 100°C, %

20

20 30 40 50 60 70 80 90 100100

40

60

80

100

120

0Pe

rmea

bilit

y, m

D

Fluid volume, pore volumes

2% KCI (injection direction)2% KCI (production direction)Fibers injected with spent acid (pH = 6.5)

16-hshut-in

K0 K1

K2K3

K4K5

K6K7

N2

16-hshut-in

10

9

8

7

6

5

4

3

2

1

00 5 10 15 20 25 35 45 50 5530 40

In addition, the model predicts fluid diversion behavior (previous page, top).

After demonstrating the diversion capabili-ties of fiber-laden VDA fluids in the laboratory, the developers considered the effects of fibers on reservoir productivity following an acidizing treatment. If fibers remained in the wormholes indefinitely, their presence would hinder the flow of fluids from the reservoir to the wellbore. For this reason, degradable fibers were viewed as an attractive option. After a treatment, the fibers hydrolyze and degrade within a few days. The absence of fibers leaves unobstructed wormholes and maximizes formation productivity. Further-more, the degradable fibers are composed of an organic acid polymer whose degradation prod-ucts are acidic, giving rise to further formation stimulation (right).9

The results of the laboratory study were suffi-ciently encouraging to allow the engineers to advance to the next development stage—yard testing to demonstrate that the fiber-laden MaxCO3 Acid fluid could be prepared and pumped efficiently and safely.

Verifying Wellsite DeliverabilityBecause matrix acidizing treatments typically consume small fluid volumes compared with other stimulation techniques, engineers usually employ batch-mixing procedures. By contrast, fracture acidizing usually requires large fluid vol-umes, and continuous mixing is necessary to keep pace with the higher pump rates. Consequently, engineers needed to develop methods for mixing MaxCO3 Acid formulations in both scenarios. The principal objectives were to disperse the fibers safely and efficiently in the fluid and prepare a uniform suspension. Because the degradable fibers are light and finely divided, engineers were challenged to devise ways to immerse the fibers in the VDA fluid so that they would form a homogeneous mixture.

Experimentation led to the discovery that uniform MaxCO3 Acid mixtures can be efficiently batch mixed with existing equipment (previous page, bottom). The equipment consists of a ves-sel, into which engineers pour the base VDA fluid, and an 8,000-L [50-bbl] recirculating mixing tank equipped with rotating paddles. Field personnel dispense the fibers manually. Until the treatment commences, continuous agitation prevents fiber and fluid separation.

The POD programmable optimum density blender is standard Schlumberger equipment for continuously dispensing solid materials such as proppant into fracturing fluids, and it proved to

be an efficient system for preparing MaxCO3 Acid mixtures. However, the fluid exit points must be secure to ensure that personnel are shielded against fluid leaks and sprays. Therefore, engi-neers designed a special splash protection kit

12 Oilfield Review

10. Bullheading is the pumping of fluids into a wellbore from the surface with no direct control over which intervals will accept the fluids.

11. Thabet S, Brady M, Parsons C, Byrne S, Voropaev S, Lesko T, Tardy P, Cohen C and Mchaweh A: “Changing the Game in the Stimulation of Thick Carbonate Gas Reservoirs,” paper IPTC 13097, presented at the International Petroleum Technology Conference, Doha, Qatar, December 7–9, 2009.

that includes a berm below the blender and a plastic sidewall (above left). They also developed a special chute for metering the degradable fibers as they are dispersed into the mixing tub. The modified chute, mounted directly above the mixing tub, has no restrictions or bends that might hinder smooth fiber delivery.

After verifying that MaxCO3 Acid fluids could be prepared reliably with existing field equip-ment, the project team traveled to Qatar for field testing. A principal test objective was to evaluate the accuracy of the acid placement and diversion simulator.

Field Testing in QatarThe North field in Qatar is an offshore gas pro-ducer that presents unique challenges for com-pletion and stimulation (above right). The reservoir is 1,000 to 1,300 ft [300 to 400 m] thick and the wells, which may be deviated by as much as 55°, can be as long as 2,000 ft [610 m]. The res-ervoir comprises alternating sequences of lime-

> Continuous mixing of MaxCO3 Acid fluid. A POD blender is outfitted with a special fiber delivery feeder (top right) that has no restrictions or bends, thus ensuring smooth metering. Field workers place a berm (top left) under the blender to guard against fluid spills. A plastic sidewall around the mixing tubs (bottom) further shields the mixing process.

Fiber feeder

> Qatar North field. Discovered in the 1970s, this accumulation is the largest gas field in the world, with estimated reserves as high as 25.5 trillion m3 [900 Tcf]. The reservoir is called the South Pars field on the Iranian side of the maritime border (dashed black line). The producing formation is characterized by large interzonal permeability contrasts—up to a ratio of 100:1. The reservoir depth is about 3,000 m [9,800 ft] below the seabed, and the elevated hydrostatic pressure tends to favor stimulation of bottom zones at the expense of upper reservoir layers, further increasing the difficulty of achieving uniform stimulation in one treatment.

IRAN

QATAR

BAHRAINNorthField

SouthPars

SAUDIARABIA

0 km

0 mi 50

50

SAUDIARABIA

IRAN

> Jujo-Tecominoacán field. This region is among the most prolific oil and gas producing areas in southern Mexico. The reservoirs are naturally fractured and difficult to stimulate uniformly.

Villahermosa

TabascoState

Jujo-Tecominoacán Field

50

km0 50

miles0

UNITED STATES

MEXICO

Autumn 2013 13

stone and dolomite that have a permeability contrast ratio as high as 100:1.

The typical workflow for designing and per-forming a MaxCO3 Acid treatment consisted of several steps. To build a reservoir model, engi-neers first acquired a thorough description of the candidate well. The description included well completion diagrams, petrophysical and pressure log measurements and pretreatment well pro-duction data. The simulator produced a pumping schedule designed to provide optimal zonal cov-erage and maximize posttreatment reservoir per-meability. During the treatment, engineers measured the bottomhole and wellhead pres-sures and compared the results with those pre-dicted by the simulator. Posttreatment activities included production logging to further verify the accuracy of the simulator.

One test well had 290 ft [88 m] of perforations along 830 ft [250 m] of measured depth—between 12,270 and 13,100 ft [3,740 and 3,990 m]. The principal obstacles to effective acid place-ment were the high permeability contrast and hydrostatic pressure effects favoring preferential stimulation of deeper high-permeability zones (right). Prior to these field tests, installation of bridge plugs had been the preferred technique to achieve fluid diversion.

Schlumberger engineers performed a matrix acidizing treatment from a stimulation vessel using the bullheading technique.10 The treatment consisted of alternating stages of 290 bbl [46 m3] of 28% HCl and 320 bbl [51 m3] of MaxCO3 Acid fluid containing 75 lbm/1,000 galUS [9.0 kg/m3] of degradable fibers. To ensure uniform fiber sus-pension, engineers set up the treatment so that 160-bbl [25-m3] spacers of VDA fluid preceded and followed the MaxCO3 Acid stages. During the treatment, the simulated and measured bottom-hole pressures were in good agreement, provid-ing confirmation that the diversion physics of MaxCO3 Acid behavior were well described by the simulator (right).

After the success of the first test well, engi-neers performed 10 more acidizing treatments in the field with similar results.11 The fiber-laden acid performed as predicted, and operational efficiencies were gained by not having to rely on mechanical diversion. The time required to com-plete, perforate, stimulate and clean up the MaxCO3 Acid wells was two to four days shorter than that of the traditional approach, represent-ing a savings of US$ 480,000 to US$ 960,000 per well. Environmental benefits included a 72% reduction in the emission of greenhouse gases because of reduced flaring. Following the success of the Qatari field tests, the operator deployed MaxCO3 Acid technology in other regions.

> Permeability profile. The permeability varies four orders of magnitude in a test well in the Qatar North field.

Mea

sure

d de

pth,

ft

Permeability, mD

13,2000.1 1 10 100 1,000

13,100

13,000

12,900

12,800

12,700

12,600

12,500

12,400

12,300

12,200

> Simulated and measured pressures from a field test in the Qatar North field. Engineers pumped four stages of 28% HCl and MaxCO3 Acid fluid. A VDA fluid spacer preceded and followed each MaxCO3 Acid stage to preserve fiber suspension uniformity. The excellent agreement between the measured (blue curve) and simulated (black) bottomhole pressures (BHP) helped confirm the validity of the MaxCO3 Acid placement model.

6,500

7,500

6,000

7,000

8,000

5,500

5,00080 100 120 140 160

25

35

30

40

20

15

10

50

BHP,

psi

Time, min

Pum

p ra

te, b

bl/m

in

Measured BHPSimulated BHPPump rate

Fluid at perforationsMaxCO3 Acid fluidWater

GasHCIVDA acid

Optimizing Production in Southern MexicoThe Jujo-Tecominoacán field, operated by Petróleos Mexicanos (PEMEX), is located 60 km [40 mi] from Villahermosa, Tabasco, in southern

Mexico (previous page, bottom). The field has 48 producing wells and 19 injection wells to maintain reservoir pressure. The average depth of the producing intervals is 5,000 m [16,400 ft],

14 Oilfield Review

and the reservoir temperature varies between 120°C and 160°C [250°F and 320°F]. Wells in this field typically produce from multiple perforated intervals with a highly variable natural fracture density. This scenario creates a large permeabil-ity contrast between intervals that can reach 1,000:1. Consequently, achieving uniform zonal

coverage during stimulation treatment presents a major challenge.

One typical well that was drilled in 2005 has two producing intervals: from 5,274 to 5,294 m [17,303 to 17,369 ft] and from 5,308 to 5,340 m [17,415 to 17,520 ft]. The reservoir temperature and pressure are 137°C [279°F] and 22.8 MPa

[3,300 psi]. Porosity varies between 5% and 8%. The permeabilities of the upper and lower inter-vals are 1,000 mD and 3 mD; therefore, the per-meability contrast is 333:1.

The initial oil production rate was 1,278 bbl/d [203 m3/d]. Between 2006 and 2009, PEMEX per-formed several stimulation treatments using con-ventional acids and diversion techniques. The production rate increased immediately after each treatment but failed to stabilize and contin-ued to decline. In 2009, PEMEX engineers decided to evaluate the MaxCO3 Acid technology in the hope of achieving uniform and long-lasting stimulation of the two intervals.12

Schlumberger engineers performed a matrix acidizing treatment consisting of bullheading 30 m3 [7,800 galUS] of aromatic solvent preflush to clean the perforations, 60 m3 [15,600 galUS] of HCl–formic acid blend, 10 m3 [2,600 galUS] of MaxCO3 Acid fluid containing 90 lbm/1,000 galUS [11 kg/m3] fibers and 2 m3 [520 galUS] of ammo-nium chloride brine spacer (above left). Pump rates varied between 8.2 and 15 bbl/min [1.3 and 2.4 m3/min]. The last treatment stage contained nitrogen to energize the fluid and accelerate well cleanup, and hydrocarbon production commenced within three days. The initial oil production rate, 3,000 bbl/d [480 m3/d], exceeded PEMEX’s fore-cast. After three months, the average oil produc-tion rate had stabilized at 1,600 bbl/d [250 m3/d] (below left). Following the success of this treatment, PEMEX has continued to apply MaxCO3 Acid tech-nology in this field with favorable results.

> Pumping schedule for a matrix acidizing treatment in the Jujo-Tecominoacán field. During the 11-stage treatment, engineers pumped an aromatic solvent to clean up perforations, an HCl–formic acid blend, MaxCO3 Acid fluid and an ammonium chloride brine spacer. The final stage contained nitrogen [N2] to enhance well cleanup.

Fluid NameStage Name Stage FluidVolume, m3

Nitrogen Pump Rate, m3/min

Spacer 3% NH4Cl brine

Spacer 3% NH4Cl brine

Diverter MaxCO3 Acid fluid

Diverter MaxCO3 Acid fluid

Acid HCI–formic acid blend

HCI–formic acid blend

HCI–formic acid blend

Acid

Preflush Aromatic solvent

Preflush Aromatic solvent

Preflush Aromatic solvent

Acid

1

1

5

5

20

20

10

10

10

20

Flush Nitrogen

80

80

150

> Production history in a PEMEX well in the Jujo-Tecominoacán field. Initial oil production was 1,278 bbl/d [203 m3/d]. Subsequent matrix acidizing treatments employing conventional techniques failed to achieve sustained production improvements. After a MaxCO3 Acid treatment in December 2009, oil production increased to 3,000 bbl/d and stabilized at 1,600 bbl/d, exceeding the original production rate.

Oil p

rodu

ctio

n ra

te, b

bl/d

Date

Begin MaxCO3 Acid treatment

Oil production

Jan 2009 Jan 2010Apr 2009 Apr 2010July 2009 Oct 2009

2,000

2,500

3,000

3,500

1,500

1,000

500

0

12. Martin F, Quevedo M, Tellez F, Garcia A, Resendiz T, Jimenez Bueno O and Ramirez G: “Fiber-Assisted Self-Diverting Acid Brings a New Perspective to Hot, Deep Carbonate Reservoir Stimulation in Mexico,” paper SPE 138910, presented at the SPE Latin American and Caribbean Petroleum Engineering Conference, Lima, Peru, December 1–3, 2010.

13. Rahim Z, Al-Anazi HA, Al-Kanaan AA and Aziz AAA: “Successful Exploitation of the Khuff-B Low Permeability Gas Condensate Reservoir Through Optimized Development Strategy,” Saudi Aramco Journal of Technology (Winter 2010): 26–33.

14. Aviles I, Baihly J and Liu GH: “Multistage Stimulation in Liquid-Rich Unconventional Formations,” Oilfield Review 25, no. 2 (Summer 2013): 26–33.

15. Jauregui JL, Malik AR, Solares JR, Nunez Garcia W, Bukovac T, Sinosic B and Gürmen MN: “Successful Application of Novel Fiber Laden Self-Diverting Acid System During Fracturing Operations of Naturally Fractured Carbonates in Saudi Arabia,” paper SPE 142512, presented at the SPE Middle East Oil and Gas Show and Conference, Manama, Bahrain, September 25–28, 2011.

Autumn 2013 15

> South Ghawar field in eastern Saudi Arabia. The producing reservoirs, in the Khuff Formation, are composed of heterogeneous carbonates. The permeability and porosity vary widely within 100 to 200 ft [30 to 60 m] of formation thickness, presenting difficult fluid diversion challenges.

IRAN

BAHRAIN

QATAR

UNITED ARABEMIRATES

SAUDI ARABIA

South Ghawar Field

0 km

0 mi 100

100

GasOil

SAUDIARABIA

EGYPT

IRAN

Improving Gas Production in Saudi ArabiaThe vast carbonate reservoirs of Saudi Arabia are prime locations for stimulation treatments using acidic fluid systems. From simple acid washes to major acid fracturing operations, every carbon-ate stimulation technology has found an applica-tion in this region.

Most gas production in Saudi Arabia comes from the Khuff Formation, located in the eastern part of the country (right). The Khuff Formation is highly heterogeneous, exhibiting wide varia-tions in formation permeability (0.5 mD to 10 mD) and porosity (5% to 15%). It is composed mainly of calcite and dolomite interbedded with streaks of anhydrite. The average temperature and pressure are 280°F [138°C] and 7,500 psi [52 MPa].13

Saudi Aramco engineers applied MaxCO3 Acid technology during several matrix acidizing treatments, all of which yielded excellent results. Following this success, Saudi Aramco engineers decided to perform 25 acid fracturing treatments employing the MaxCO3 Acid formu-lation. Eight acid fracturing stages were per-formed in three wells equipped with openhole multistage fracturing completions that enabled continuous treatments.14 The remainder of the jobs, single-stage treatments in vertical or devi-ated wells, were completed with cemented and perforated liners.15

Engineers performed one treatment in a cemented and perforated well that had a 65° deviation. Three pay zones existed along a 240-ft [73-m] interval in the central sector of the field. From reservoir parameters obtained from open-hole logs, engineers concluded that, to meet Saudi Aramco’s production expectations, it would be necessary to pump a treatment that stimu-lated all three perforated zones simultaneously.

Engineers developed a fracturing treatment that consisted of 19 fluid stages that alternated portions of a 35-lbm/1,000 galUS [4.2-kg/m3] borate crosslinked guar fracturing fluid, 28% SXE superX emulsified acid to retard the rate of acid consumption, 28% HCl and 15% MaxCO3 Acid for-mulation with degradable fiber concentrations between 75 and 175 lbm/1,000 galUS [9 and 21 kg/m3] (right). During the treatment, after the first MaxCO3 Acid stage contacted the formation, engineers recorded a 4,500-psi [31-MPa] bottom-hole pressure rise—the first time such a large increase had been recorded in this carbonate reservoir—indicating that excellent fluid leakoff > Pumping schedule for an acid fracturing treatment in Saudi Arabia. The total fluid volume was

124,200 galUS [2,960 bbl, 470 m3], allowing simultaneous stimulation of three zones without the need for mechanical diversion techniques. Such treatment simplicity saved several days of rig time, resulting in significant operational cost savings.

Treatment Schedule

Fluid NameStage Name Stage FluidVolume, galUS [m3]

AcidConcentration, %

Pump Rate,bbl/min [m3/min]

20 [3.2]

30 [4.8]

40 [6.4]

40 [6.4]

40 [6.4]

30 [4.8]

35 [5.6]

30 [4.8]

35 [5.6]

40 [6.4]

20 [3.2]

30 [4.8]

40 [6.4]

40 [6.4]

10 [1.6]

10 [1.6]

10 [1.6]

10 [1.6]

40 [6.4]

0

0

0

0

0

0

0

15

15

15

28

28

28

28

0

0

15

28

0

Pad

Pad

Pad

Pad

Pad

Pad

Pad

Diverter 1

Diverter 2

Diverter 3

Acid 1

Acid 2

Acid 3

Acid 3

Overflush 2

Flush

Diverter 4

Acid 4

Overflush 1

Crosslinked 35-lbm gel

Crosslinked 35-lbm gel

Crosslinked 35-lbm gel

Crosslinked 35-lbm gel

Crosslinked 35-lbm gel

Crosslinked 35-lbm gel

Crosslinked 35-lbm gel

MaxCO3 Acid fluid

MaxCO3 Acid fluid

MaxCO3 Acid fluid

SXE emulsified acid

SXE emulsified acid

SXE emulsified acid

SXE emulsified acid

Overflush

Water

MaxCO3 Acid fluid

28% HCl

Overflush

9,000 [34]

9,000 [34]

9,000 [34]

3,000 [11]

10,000 [38]

3,000 [11]

3,000 [11]

3,000 [11]

3,000 [11]

9,000 [34]

9,000 [34]

9,000 [34]

9,000 [34]

5,000 [19]

11,200 [42]

3,000 [11]

7,000 [26]

7,000 [26]

3,000 [11]

16 Oilfield Review

> Pressure and temperature data. During a Saudi Aramco acid fracturing treatment, the pumping rate (blue line) varied from 10 to 40 bbl/min [1.6 to 6.4 m3/min], and the bottomhole treating pressure (red line) exceeded the formation fracturing pressure (dashed black line) throughout most of the treatment. The vertical blue bars denote periods during which MaxCO3 Acid fluid entered the perforations.

8,000

6,600

5,200

3,800

2,400

1,000

10

10 30 50 70 90 110 130 150 170

25

40

55

70

85

100

115

9,400

10,800

12,200

15,000

13,600

Pres

sure

, psi

Treatment time, min

Fracturing pressure

Rate

, bbl

/min

10

1Bottomhole treating pressurePump rate

> The presalt reservoirs of Brazil. The main producing fields are located primarily offshore (left). The reservoirs are in carbonate formations that lie underneath a thick layer of evaporite minerals (right). The reservoir depth is between 4,500 and 6,500 m [14,800 and 21,300 ft].

BRAZIL

Salt

Dept

h, m

0

1,000

2,000

3,000

4,000

5,000

6,000

7,000

8,000

9,000

Overburden formations

Presaltoil

Rio de Janeiro

Espirito SantoBasin

Campos Basin

Santos Basin

São Paulo

Curitiba

SOUTH AMERICA

km 5000

mi 5000

control and diversion had been achieved (left). Moreover, the bottomhole pressure exceeded the fracturing pressure throughout most of the treat-ment, which had not been possible to achieve during previous attempts using conventional diversion techniques.

After the treatment, the well cleaned up in less than three days; previously, four to five days had been necessary. Prior to the treatment, the gas production rate had been 8 MMcf/d [230,000 m3/d] with a wellhead pressure of 2,060 psi [14.2 MPa]. The posttreatment produc-tion rate was 23 MMcf/d [650,000 m3/d]—a nearly threefold increase—with a wellhead pres-sure of 2,230 psi [15.4 MPa]. The excellent post-stimulation performance of this well has been observed in the majority of other wells in this region treated with the fiber-laden acid.

Elimination of mechanical diversion tech-niques reduced the well completion and stimula-tion time up to six days, resulting in a savings of US$ 480,000 to US$ 600,000. As a result, the MaxCO3 Acid system is now a prominent element of Saudi Aramco’s stimulation strategy.

Autumn 2013 17

>Matrix acidizing treatment. In a presalt well offshore Brazil, engineers pumped 13 fluid stages consisting of alternating portions of 15% HCl, VDA diverter and MaxCO3 Acid fluid at various pump rates (blue curve). A mixture of 15% HCl and a mutual solvent preceded and followed the treatment. As the treatment progressed, the rig pressure (red curve) and bottomhole pressure (green curve) rose, indicating that the fibers were effectively diverting treatment fluid to zones with lower permeability.

0 1,00000

1,000

2,000

3,000

4,000

5,000

6,000

7,000

8,000

4

8

12

16

20

24

28

32

36

40

2,000 3,000

Time, s4,000

4,000

4,500

5,500

6,500

7,500

5,000

6,000

7,000

8,000

5,000 6,000 7,000 8,000 9,000 10,000

Pum

p ra

te, b

bl/m

in

Rig

pres

sure

, psi

Botto

mho

le p

ress

ure,

psi

HCl plus mutual solvent15% HClVDA fluidMaxCO3 Acid fluid

Stimulating Oil Production in Offshore BrazilIn South America, the presalt region comprises a group of oil-bearing carbonate formations located in an offshore region along the coast of Brazil (previous page, bottom).16 The produc-ing formations occur at depths between about 4,500 and 6,500 m [14,800 and 21,300 ft] and lie directly underneath a 2,000-m [6,500-ft] layer of evaporite minerals. The reservoir tem-peratures vary between about 60°C and 133°C [140°F and 272°F].

The producing carbonate reservoir is a result of the deposition of mollusks followed by diagen-esis. Such reservoirs, called “coquinas,” feature large variations in reservoir properties. Porosity varies from 5% to 18%, and permeability varies from less than 0.001 mD to tens of mDs. Such het-erogeneity presents an especially difficult diver-sion challenge during stimulation treatments.

Engineers at Petrobras decided to evaluate the MaxCO3 Acid fiber-assisted diversion tech-nology in a new well in the Pirambu field. Using the acid placement and diversion simulator, Schlumberger engineers designed a matrix acidizing treatment for an interval between 4,500 m and 4,570 m [14,800 and 15,000 ft]. The simulator called for a 790-bbl [12.6-m3], 13-stage bullheaded treatment consisting of alternating volumes of 15% HCl, VDA fluid and MaxCO3 Acid fluid with a fiber concentration between 100 and 120 lbm/1,000 galUS [12 and 14 kg/m3]. The treatment was preceded by a brine and HCl mixture containing a monobutyl ether mutual solvent.17 After the treatment, engineers pumped another volume of HCl with mutual solvent followed by diesel to accelerate well cleanup. The pump rate varied from 5 bbl/min [0.8 m3/min] during the MaxCO3 Acid fluid stages to 10 bbl/min [1.6 m3/min] during the injection of HCl and to 20 bbl/min [3.2 m3/min] during the VDA diverter stages (above).

After well cleanup, engineers at Petrobras evaluated the results by performing production logging. The logs showed that the well was pro-ducing from all of the treated zones as pre-dicted by the simulator. Since this treatment,

Petrobras has continued to specify the use of MaxCO3 Acid fluid.

Refining MaxCO3 Acid TechnologyAs of this writing, more than 300 MaxCO3 Acid stimulation treatments have been performed around the world. In addition to the examples featured in this article, treatments have been performed in Kazakhstan, Angola, Canada, the US, Kuwait and the Caspian Sea.

As the number of treatments has increased, the larger treatment database has allowed con-tinuous refinement of the simulator and improve-ment of stimulation results in naturally fractured carbonate reservoirs. The technique has also allowed operators to reduce or eliminate the use of ball sealers or packers, thereby reducing costs and operational risks.

At present, work is underway to combine MaxCO3 Acid technology with the ACTive family of live downhole coiled tubing services. This arrange-ment employs distributed temperature sensors that will allow engineers to monitor fluid place-ment in real time and change treatment designs during a job. Such flexibility will further enhance the effectiveness of acidizing treatments employ-ing fiber-based fluid diversion. —EBN

16. Beasley CJ, Fiduk JC, Bize E, Boyd A, Frydman M, Zerilli A, Dribus JR, Moreira JLP and Pinto ACC: “Brazil’s Presalt Play,” Oilfield Review 22, no. 3 (Autumn 2010): 28–37.

17. Mutual solvents are chemicals in which both aqueous and nonaqueous compounds are miscible. These solvents may be used to prevent emulsions, reduce surface tension and leave formation surfaces water-wet.

18 Oilfield Review

Casing Corrosion Measurement to Extend Asset Life

Corrosion challenges are not new to the oil and gas industry, and producers are

continually seeking new ways to keep corrosion at bay. Experts have made advances

in corrosion monitoring along several fronts. The implementation of these technolo-

gies may help operators optimize infrastructure utilization, maximize production and

minimize negative impact on the environment.

Dalia AbdallahMohamed FahimAbu Dhabi Company for Onshore Oil OperationsAbu Dhabi, UAE

Khaled Al-HendiMohannad Al-MuhailanRam JawaleKuwait Oil CompanyAhmadi, Kuwait

Adel Abdulla Al-KhalafQatar PetroleumDoha, Qatar

Zaid Al-KindiAbu Dhabi, UAE

Abdulmohsen S. Al-Kuait Hassan B. Al-Qahtani Karam S. Al-Yateem Saudi AramcoDhahran, Saudi Arabia

Nausha AsrarSugar Land, Texas, USA

Syed Aamir Aziz J.J. KohringDhahran, Saudi Arabia

Abderrahmane BenslimaniAhmadi, Kuwait

M. Aiman FituriDoha, Qatar

Mahmut SengulHouston, Texas

Oilfield Review Autumn 2013: 25, no. 3. Copyright © 2013 Schlumberger.For help in preparation of this article, thanks to Ram Sunder Kalyanaraman, Clamart, France.Avocet, EM Pipe Scanner, FloView, Petrel, PipeView, PS Platform, Techlog, UCI and USI are marks of Schlumberger.

Oil and gas companies typically serve two mas-ters. On the one hand, profitability dictates that producers maximize long-term production while minimizing operating expenditures. On the other

hand, environmental compliance requires that companies conduct exploration and production operations safely and in an environmentally responsible manner.

> Typical refining-corrosion life cycle for metals. Energy is stored in a metal as it is refined from its naturally occurring state (such as iron ore) to an alloy. Corrosion takes place spontaneously and releases the stored energy, which returns the metal back to a lower energy state. That process can be slowed by the application of one or more field-based mitigation measures.

Energy Added During Refining

Refined Metal or AlloyIron Ore (Oxides) and Corrosion Products

Energy Released by Corrosion

Autumn 2013 1919

The two mandates share a common enemy. Corrosion, which is the natural tendency for materials to return to their most thermodynami-cally stable state by reacting with agents in the surrounding environment, attacks almost every component of a well. Wells are constructed pri-marily of steel, which is refined from naturally occurring iron ore. The process of refining ore into a steel alloy suitable for oil and gas drilling and production takes the ore to a higher energy state. Corrosion reverses this process and brings metal back toward its original, lower energy state (previous page).1

The process of corrosion, which begins the moment steel is cast, is accelerated in the oil field by the presence of acidic species—such as hydrogen sulfide [H2S] or carbon dioxide [CO2]—in many formation fluids and by the elevated temperatures and pressures in producing forma-tions. The consequences of corrosion include a reduction in wall thickness and loss of strength, ductility and impact strength in the steel that makes up the downhole tubulars, wellheads and surface piping and downstream processing equip-ment (right).

Failure to address corrosive attacks early impacts well profitability because operators must then implement potentially expensive, and per-haps extensive, mitigation methods. Not only does mitigation increase operating expenses, it may force operators to shut a well in for some period of time. In the worst cases, unattended corrosion can lead to a leak or rupture, which may threaten the safety of oilfield personnel, lead to production losses and introduce hydrocarbons and other reservoir fluids into the environment.

The total annual cost of corrosion in the US alone is estimated at approximately US$ 1.4 bil-lion, of which US$ 589 million is surface pipeline and facility costs, US$ 463 million is downhole tubing expenses and US$ 320 million is capital expenditures.2 These estimates do not factor in the fines that may be levied by government regu-latory agencies against operators that experience a corrosion-related discharge of production fluids into the environment. The costs and risks may also increase as hydrocarbon sources are discovered in more-challenging environments—deeper reservoirs with higher temperatures and pressures that contain higher concentrations of

1. For more on the corrosion process: Brondel D, Edwards R, Hayman A, Hill D, Mehta S and Semerad T: “Corrosion in the Oil Industry,” Oilfield Review 6, no. 2 (April 1994): 4–18.

2. Koch GH, Brongers MPH, Thompson NG, Virmani YP and Payer JH: “Corrosion Costs and Preventive Strategies in the United States,” Washington, DC: US Department of Transportation Federal Highway Administration, Office of Infrastructure Research and Development, Publication no. FHWA-RD-01-156, September 2001.

> Summary of corrosion problems and solutions. In the oil field, corrosion is pervasive and takes many forms. By properly identifying the source of corrosive attack, an operator can implement a suitable corrosion monitoring and control program.

Problem Control MethodsCause of Corrosion Monitoring

Oxygen corrosion

Sulfate-reducingbacteria (SRB)

Hydrogen sulfide stress corrosion crackingHydrogen-inducedcracking

Acid corrosion

Galvanic (bimetallic) corrosion

Pitting corrosion (rapidcorrosion at defectsin inert surface film)

Subdepositcorrosion

Chloride corrosion(rapid cracking onexposure to hotchloride media)

Fatigue

Crevice corrosion

Hydrogen sulfide corrosion pitting

Carbon dioxidecorrosion

• Water and oxygen sampling• Iron counts• Corrosion probes• Oxygen sensors• Coupon surveys• Wall thickness surveys• Visual internal inspections• Visual surveys

• Anaerobic bacteria counts• Chlorine residuals measurements

• Materials quality control

• Acid inhibitor checks

• Design reviews

• Equipment inspections

• Equipment inspections• Bacteria counts

• Equipment inspections• Oxygen analyses

• Equipment inspections

• Equipment disassembly and inspections• Leak detections

• Probes• Iron counts• Wall thickness surveys

• Probes• Iron counts• Wall thickness surveys

• Resistant materials• Oxygen scavengers• Oxygen stripping• Improved seal design• Coatings• Cathodic protection

• Biocides• Chlorination

• Suitable materials

• Acid inhibitors

• Improved design

• Electrical isolation of metals (cathodic coating)

• Materials selection

• Pigging• Biocides• Improved sealing and design• Minimum velocity design

• Materials selection

• Vibration design

• Improved design • Materials selection

• Control of contaminated gas

• Degassing at low pressures

• Use of resistant materials

• Degassing at low pressures• Control of contaminated gas• Use of resistant materials

• Oxygenated water• Internal attack• External attack

• Anaerobic fluids• Stagnant fluids• Conditions under scales or other deposits

• Produced fluids containing hydrogen sulfide• Anaerobic systems contaminated with SRB

• Stimulation and cleaning acids

• Two metals with different ionic potentials in a corrosive medium

• Immersion• Inert surface films

• Wet solids deposits• Biofilms• Porous gaskets

• Salt solution• Oxygen and heat

• Rotating equipment• Wave-, wind- or current-induced loading

• Poor design• Imperfections in metal

• Water from production aquifer or other deep aquifer• Water contaminated by stripping or lift gas

• Water from production aquifer or other deep aquifer• Water contaminated by stripping or lift gas

20 Oilfield Review

acidic gases—which may present more-aggres-sive corrosion environments.

The industry has advanced several methods to combat corrosion and extend the operating life of a well. These may be broadly classified into four main categories: • metallurgy—substituting traditional wellbore

tubulars with those manufactured with a corro-sion-resistant alloy (CRA)

• chemical—modifying production fluids to reduce the intensity of corrosive attacks or cre-ating barriers that isolate the metal from pro-duced fluids through the application of a protective coating

• injection—pumping surfactant-base fluids that aggregate at the metal surface and block metal-water contact, thus inhibiting corrosion

• cathodic protection—using DC current to cre-ate impressed cathodic protection.3

The first option—upgrading tubulars to those composed of CRA—may be cost prohibitive on a large scale. In the US alone, there are more than 100,000 producing oil and gas wells with casing, tubing, wellheads, processing equipment and gathering lines.

Manufacturers may employ another mitiga-tion option: applying permanent coatings, which combat corrosion by forming a resistant barrier between the corrosive fluid media and the metal surface. Many coating types exist and are gener-ally categorized as follows:• metallic—zinc, chromium and aluminum• inorganic—enamels, glasses, ceramics and

glass-reinforced linings• organic—epoxies, acrylics and polyurethanes.4

As with CRAs, coatings may promise a longer operating life with reduced maintenance, but they come at a cost premium.5

Operators may use inhibition by chemical means during the production stage of the well to mitigate corrosion on the internal surface of pip-ing and equipment. Corrosion inhibitors are typi-cally surfactant-base chemical formulations that are added to the production stream in concentra-tions ranging from tens to several hundred parts per million (ppm). The inhibitor molecules migrate and collect at surfaces; in the case of a well’s production infrastructure, the molecules collect at the metal surface to form a barrier between it and the corrosive fluid phase. In this way, they act in a manner similar to that of a coat-ing, but at a lower cost than that of a permanent coating or a CRA. Unlike a coating, a corrosion inhibitor must be reapplied to replenish the inhibitor film that is degraded or washed away by the flowing action of the production stream.6

Corrosion prevention through cathodic pro-tection works by forcing anodic areas of the metal—those susceptible to corrosive attack—to become cathodic or noncorrosive. To accom-plish this, operators apply a DC current through the metal to counteract the corrosion current—a technique known as impressed cathodic protec-tion (ICP)—or use sacrificial anodes, which are composed of metal that has a greater corrosion tendency than the metal to be protected.7

This article focuses on corrosion monitoring and measurement techniques for downhole infra-structure during production. Case studies from

the Middle East demonstrate how corrosion mon-itoring tools and mitigation technologies have helped operators identify the location and sever-ity of corrosion in the subsurface infrastructure, which informed each company’s choice of mitiga-tion solution.

Corrosion and the Life CycleCorrosion is a major concern throughout the life of a well, and specific considerations and mitiga-tion strategies are required at each stage. Asset personnel usually begin making corrosion miti-gation decisions for a well before drilling. During the well design stage, the operator con-ducts comprehensive reservoir studies, which include reservoir simulation modeling, core studies and fluid analysis from offset well data. Engineers use the information obtained from these studies to develop risk assessments for corrosion threats in subsequent stages of the well. Engineers then develop and implement mitigation strategies that include appropriate materials selection, optimal production rates, monitoring programs and corrosion inhibitor treatments (above).

During the drilling process, operators focus corrosion mitigation strategies on extending the working life of drillpipe, which is exposed to high operational stresses as well as potentially corro-sive drilling muds and formation fluids. The drill-pipe may undergo one of several types of corrosion mechanisms, including localized pitting, in which H2S, chloride salts or oxygen in water-base drill-ing muds cause a corrosion rate that exceeds 25 cm [9.8 in.] per year.8 Other corrosion sources

> Corrosion considerations at each stage of the asset life cycle. During each stage of a well’s life, engineers must consider operational factors to keep corrosion at bay and minimize the threat of production fluid leaks into the surrounding environment.

• Perform reservoir modeling.

• Perform core analysis.

• Perform materials selection.

• Perform risk analysis.

WellDesign

WellDecline Decommissioning

Drilling andCompletions Production

• Select suitable drilling mud.

• Select suitable alloys for pipe work and equipment.• Select suitable oxygen and sulfide scavengers.

• Ensure long-term containment of abandoned well.

• Ensure compliance with environmental regulations.

• Implement more-stringent and expansive asset integrity evaluation. • Implement or expand oil and water separation operations.

• Use corrosion monitoring tools and services.

• Use corrosion mitigation technologies (corrosion inhibitors, sand control systems and oxygen scavengers).• Evaluate infrastructure condition and track corrosion rates. • Implement repairs and replacement strategies as needed.

Prod

uctio

n

Autumn 2013 21

include the presence of CO2 at a partial pressure of 20 to 200 kPa [3 to 30 psi] or greater, microbio-logically influenced corrosion (MIC) caused by the presence of certain bacteria (microbes) in produced fluids and crevice corrosion in which localized corrosion rates at metal-to-metal or metal-to-nonmetal interfaces, such as at joint couplings or gaskets, reach elevated levels and lead to pitting or cracking.9

The common ingredient in these various cor-rosion events is drilling mud. To prevent drilling muds from becoming corrosive, mud engineers use specific chemical treatments in the mud. These treatments focus on keeping the pH of the mud within an acceptable range—typically between 9.5 and 12—by dosing it with alkali or adding oxygen scavengers to reduce dissolved oxy-gen levels below 1 ppm or by adding sulfide scav-engers that eliminate H2S from the mud system.10

The completion phase of a well refers to the assembly and installation of downhole tubulars and equipment such as packers and artificial lift pump systems. Information collected during the well planning stage, including the temperature and pressure of the reservoir and the composition of the production fluids, helps inform the opera-tor’s decision on corrosion mitigation measures to be included in the completion. For example, anticipation of H2S or CO2 production may lead the operator to use CRAs in the completion casing strings, control valves, permanent downhole gauges and hydraulic and electric control lines.11

At the end of the well’s life cycle, hydrocarbon production levels fall—often with a correspond-ing rise in water production rates—to a point at which the well is no longer profitable and the operator must plug and abandon (P&A) it. The operator’s corrosion mitigation strategies shift to permanently prevent reservoir fluid releases to the environment long after the well is aban-doned. The basics of a P&A operation include removing completion hardware, setting isolation plugs and squeezing cement into the annular spaces at various depths to permanently seal off producing and water-bearing zones.12

P&A operations represent a pure cost, which motivates operators to conduct these activities as quickly and efficiently as possible. At the same time, a P&A job must be carried out with strict adherence to government regulatory require-ments. While these regulations vary widely in their severity and punitive measures, should a regulator find a leak in a previously abandoned well, it is the responsibility of the operator to return to make any necessary repairs and replug the well—often at a significantly higher cost than that of the original P&A operation.

Operators realize a profit during a well’s pro-duction stage, which may last from only a few years to several decades. During this stage, cor-rosion mitigation efforts are generally focused on keeping corrosion rates low and preventing leaks (below). The operator must continually

monitor and inspect the infrastructure to gauge the integrity of downhole and surface piping and equipment and the effectiveness of the mitigation.

Companies use a variety of corrosion monitor-ing techniques in oil and gas fields. Techniques are

3. Nalli K: “Corrosion and Its Mitigation in the Oil & Gas Industry—An Overview,” PetroMin Pipeliner (January–March 2010): 10–16.

4. Heim G and Schwenk W: “Coatings for Corrosion Protection,” in von Baekman W, Schwenk W and Prinz W (eds): Handbook of Cathodic Corrosion Protection, 3rd ed. Houston: Gulf Coast Publishing Company (1997): 153–178.

5. Craig BD, Lane RA and Rose DH: Corrosion Prevention and Control: A Program Management Guide for Selecting Materials, Spiral 2, 2nd ed. Rome, New York, USA: Advanced Materials, Manufacturing, and Testing Information Analysis Center, Alion Science & Technology (September 2006): 40.

6. Corrosion inhibitors are applied either continuously by strategically injecting them into the well or production string at a steady rate to maintain a desired concentration or through batch application, wherein a larger volume often called a batch, or slug, of inhibitor is applied into the well on a periodic basis. Continuous injection provides an added benefit in that the inhibitor can be applied without shutting in the well.

7. For more on impressed cathodic protection: Brondel et al, reference 1.

> Corrosion’s impact on casing integrity. Casing leaks typically arise from excessive corrosion in the production system. These leaks, which can prove costly and environmentally damaging, may allow additional formation water and sand to enter the production string of the well (blue arrow). Alternatively, crossflows (green arrows) may result, which can be difficult to characterize and treat, and in severe cases, the operator may have to pull and replace the entire casing string.

Water sand

Oil sand

Perforations

Packer

Cement sheath

Corrosion-inducedcracks

8. The corrosion rate is the thickness of metal that would be lost to corrosion in one year. This rate clearly indicates that a hole would be created in drillpipe wall in far less than a year.

9. For more on microbiologically influenced corrosion: Augustinovic Z, Birketveit O, Clements K, Freeman M, Gopi S, Ishoey T, Jackson G, Kubala G, Larsen J, Marcotte BWG, Scheie J, Skovhus TL and Sunde E: “Microbes—Oilfield Enemies or Allies?,” Oilfield Review 24, no. 2 (Summer 2012): 4–17.

10. Sloat B and Weibel J: “How Oxygen Corrosion Affects Drill Pipe,” Oil and Gas Journal 68, no. 24 (June 1970): 77–79.

11. Saldanha S: “Intelligent Wells Offer Completion Solution for Lower Tertiary Fields,” Offshore Magazine 72, no. 8 (August 1, 2012): 54–57.

12. For more on plug and abandonment operations: Abshire LW, Desai P, Mueller D, Paulsen WB, Robertson RDB and Solheim T: “Offshore Permanent Well Abandonment,” Oilfield Review 24, no. 1 (Spring 2012): 42–50.

22 Oilfield Review

selected based in part on the system’s ease of implementation for a given application or location in the production system, the ease with which results can be interpreted and the relative severity of corrosive attack. Some corrosion measurement techniques use inline monitoring tools placed directly in the production system; these tools are exposed to the flowing production stream. Other techniques provide analysis of corrosion effects after the fact in a laboratory setting.13

The weight loss technique using coupons, a direct visual identification method, is a well-known and simple monitoring method. This tech-nique exposes a specimen of material—the coupon—to the process environment for a given period of time before a technician removes it from the system and analyzes it for its physical condition and the amount of weight lost.14 The

coupon technique is advantageous because cou-pons can be fabricated from the same alloy that makes up the system under study, the corrosion rate can be easily calculated from the coupon’s weight loss over the time of exposure and the technique allows visual verification of corrosion deposits or localized corrosion. However, if a cor-rosion event such as a leak occurred while the coupon was in the system, the operator could not use the coupon alone to accurately pinpoint its time of occurrence. In addition, the coupon tech-nique is applicable only in system locations that provide easy or practical access for placing and extracting the coupon.

This second limitation makes coupon moni-toring, or any visual inspection technique, essen-tially impossible for the well’s downhole tubulars and casing strings. The remaining options are indirect measurement techniques that incorpo-

rate one or more of the various logging tools that are deployed downhole via wireline, tractor or coiled tubing.

Advances in Downhole Corrosion MonitoringLogging techniques for monitoring downhole cor-rosion include ultrasonic, electromagnetic and mechanical methods that yield detailed informa-tion about the location and extent of a corrosion event. Ultrasonic monitoring employs a central-ized sonde that is immersed in well fluid and uses a subassembly containing a rotating transducer to perform measurements.15 Most ultrasonic tools work by the principle of pulse echo measure-ment, and operators choose a transducer with the characteristics necessary for the type of mea-surement to be taken. Measurements include cement evaluation, openhole imaging and corro-sion imaging.

A USI ultrasonic imager transducer, which transmits an ultrasonic signal at a frequency ranging from 200 to 700 kHz to make the casing resonate, is typically designed for cement evalua-tion and pipe inspection. The quality of the cement bond is directly related to the degree of casing resonance: A good cement bond dampens the acoustic signal and causes a low-amplitude secondary signal to be returned to the trans-ducer; a poor cement job or free pipe allows the casing to ring and returns a higher amplitude echo. Additionally, USI measurements include 2D internal radius imaging of the casing—derived from the traveltime of the main echo from the internal surface—and the 2D casing thickness, derived from the frequency response.

Higher resolution casing measurements may be acquired with the UCI ultrasonic casing imager, which uses a focused 2-MHz transducer with improved resolution compared with that of the USI tool (left).16 The UCI tool records two echoes: the main echo from the internal surface of the casing and the smaller echo from the external surface. The radius and thickness of the casing are com-puted from the arrival times of the two echoes. The relative sizes, or amplitudes, of the two echoes are qualitative indicators of the casing condition. Although the UCI device provides a better indica-tion of the condition of the casing than does the USI imager, use of the UCI tool is limited to opera-tions in which the well fluid comprises brines, oil and light oil–base or water-base muds. Weighted muds produce an acoustic attenuation that is too strong to allow meaningful measurement.

Ultrasonic inspection provides several advan-tages as a corrosion measurement tool, including its sensitivity to both internal and external

> Basic principles of the UCI ultrasonic corrosion imager. The UCI tool uses a 2-MHz focused transducer to improve the resolution of the ultrasonic measurement. The transducer also acts as a receiver of the reflected signal and records its amplitude and time of arrival. This signal is emitted (or pulsed) through the well fluid and into the casing (top). As this signal encounters a discontinuity, such as the inner or outer wall of the casing (center), the signal is reflected back. Most of the energy is reflected in the initial echo at the inner casing wall because of the large impedance contrast between the mud and the steel; the remaining energy transmitted into the casing is again reflected at the outer wall. The signal reflected back at the inner wall can be used to evaluate the casing condition and radius. The time difference between the first two echoes can be used to determine the thickness of the casing (bottom). In comparison, the USI tool is more commonly used for ultrasonic pipe inspection and employs a 200- to 700-kHz unfocused ultrasonic transducer to induce a casing resonance. In the USI measurement, thickness is determined from the resonance frequency. (Adapted from Hayman et al, reference 15.)

Amplitude

Ampl

itude

Transducer Ultrasonic signal

Casing

Radius Thickness

Time

Autumn 2013 23

RZ

RZ

TZ

RLS

RLL

TL

RLS

RLL

TH

RP

Discriminatortransmitter, TH

Tool outerdiameter

PipePadreceiver, RP

2D thickness

2D discrimination

TL

RP RP

TH

RP

Skin depth decay

Average thickness

RLS

RLL

RLL

RLS

TL

d

Z properties

RZ

RZ

TZ

=ID0

1 1 μμτ

ωσ

defects and instantaneous in-field notification when a defect is encountered. In addition, the technique requires access to only one side of the material to gauge the condition of the entire object and obtain detailed exterior and interior images of the object. However, inspection is diffi-cult for materials that are heterogeneous in com-position, irregular in shape or thin; to improve the results of the inspection, technicians must prepare the internal surface prior to measure-ment by scraping away scale or other debris.

Operators may also employ another corro-sion monitoring method: electromagnetic (EM)-based inspection. The basic principle of this technique involves measuring the changes to a magnetic field as it passes through a metal object; the changes are related to the condition of the material such as its thickness and its electromagnetic properties.

The industry currently uses two EM corro-sion monitoring tools. The first, a flux leakage tool, magnetizes the metal object using an elec-tromagnet. When the magnetic flux encounters a damaged section or hole in the material, part of the flux leaks out of the metal; coils on the tool’s sensors detect this leakage. While this method is useful for measuring abrupt changes in pipe thickness, such as pitting or holes in the inner string, and the location of those changes, it is less effective for monitoring the steady increase of corrosion or corrosion that varies gradually over a large section of pipe or concen-tric casing configurations.

The second EM-based monitoring technol-ogy—the remote field eddy current tool—mea-sures the signal of not only the primary EM field but also the secondary field from the induced eddy currents in the surrounding pipe.17

The EM Pipe Scanner electromagnetic casing inspection tool makes four distinct measure-ments. Using a transmitter—which operates at three frequencies—and two receivers, the EM Pipe Scanner tool makes a measurement of impedance (Z), which depends on the casing’s electrical and magnetic characteristics. Using a low-frequency signal transmitter in the middle of the tool and two sets of receivers—one set above and the other below the transmitter—the tool measures the average thickness of the metal nor-malized by the skin depth.18

The remaining two measurements provide 2D images of the pipe; the tool obtains these mea-surements by pressing pad sensors against the inner wall of the pipe. One measurement uses a low-frequency signal to obtain 2D thickness information, and the other uses a high-frequency

13. “Introduction to Corrosion Monitoring,” Metal Samples: Corrosion Monitoring Systems, www.alspi.com/introduction.htm (accessed March 20, 2013).

14. “Introduction to Corrosion Monitoring,” reference 13.15. Hayman AJ, Hutin R and Wright PV: “High-Resolution

Cementation and Corrosion Imaging by Ultrasound,” Transactions of the SPWLA 32nd Annual Logging Symposium, Paris, June 16–19, 1991, paper KK.

16. Hayman AJ, Parent P, Rouault G, Zurquiyah S, Verges P, Liang K, Stanke FE and Herve P: “Developments in Corrosion Logging Using Ultrasonic Imaging,” Transactions of the SPWLA 36th Annual Logging Symposium, Paris, June 26–29, 1995, paper W.

17. For more on electromagnetic induction as a corrosion monitoring method: Acuña IA, Monsegue A, Brill TM, Graven H, Mulders F, Le Calvez J-L, Nichols EA,

discriminator transmitter located on the tool mandrel to generate signals that do not penetrate the pipe wall, creating a 2D map that discrimi-nates between damage on the inside and outside walls. Changes in the geometric properties of the metal, such as thickness or diameter, will cause changes to the mutual impedance, which is caused by flaws in the casing.

Since 2009, the EM Pipe Scanner tool has been used in wells around the world to detect

large holes, casing splits and corrosion-related metal loss from both the internal and external surfaces of casing; the tool can also measure loss from an outer casing string when multiple strings are present. The tool consists of 18 radial arms with pad sensors affixed around a slim mandrel. The sensors scan the interior surface and local thickness of production casing; the mandrel measurement helps identify average metal loss, damage and splits in the casing (above).

, EM Pipe Scanner tool. The tool (left) makes four measurements. The Z-properties measurement (bottom right) uses a transmitter (Tz) operating at three frequencies and one of the tool’s two receivers (Rz). The impedance response signal depends strongly on the dimensionless quantity, τ, which is a function of the pipe internal diameter (ID), the angular frequency ω and the electromagnetic properties of the pipe metal: the permeability μ and the conductivity σ. The term μ0 is the constant permeability of free space. The average pipe wall thickness, d, is determined using the low-frequency transmitter (TL) in the middle of the tool, along with two receivers above and two below the transmitter (center right). Two low-frequency receivers (RLL) are termed long-spacing receivers and two are termed short-spacing receivers (RLS). The phase shift of the signal—which is a function of skin depth δ—as it goes through the pipe at the transmitter and again at each receiver is used to determine the thickness of the pipe d/δ. Near the top of the tool, 18 caliper arms press pad receivers (RP) against the inside of the pipe (top right). Combining measurements from these sensors with the low-frequency signal from the transmitter (TL) at the middle of the tool provides a 2D thickness measurement. The 18 sensors are also used with a high-frequency discriminator transmitter (TH) located on the tool mandrel aligned with the sensor pads (top left). The high-frequency signal does not penetrate the pipe wall; this part of the tool provides a 2D map from signals that discriminate damage on the inside wall from signals that may indicate other phenomena.

Zapata Bermudez F, Notoadinegoro DM and Sofronov I: “Scanning for Downhole Corrosion,” Oilfield Review 22, no. 1 (Spring 2010): 42–50.

Brill TM, Le Calvez JL, Demichel C, Nichols E and Zapata Bermudez F: “Electromagnetic Casing Inspection Tool for Corrosion Evaluation,” paper IPTC 14865, presented at the International Petroleum Technology Conference, Bangkok, Thailand, February 7–9, 2012.

18. When the EM field encounters a conducting material such as the metal of a pipe, the amplitude of the field decreases exponentially with a characteristic rate given by the skin depth. A conductive and ferromagnetic material, such as casing, has a short skin depth. All media other than a vacuum have shorter skin depths at higher frequencies.

24 Oilfield Review

Operating companies can obtain these mea-surements without having to pull the completion tubing out of the hole, which saves rig time and intervention expense. While the engineer lowers the EM Pipe Scanner tool into the well on wire-line, tractor or coiled tubing, the tool conducts an initial high-speed reconnaissance run to flag areas of interest for detailed diagnostic scans to be performed as the tool is retrieved to the sur-face. The tool records a continuous log of both the average casing inner diameter and total metal thickness and provides corrosion esti-mates. The tool responds to overall metal thick-ness, allowing corrosion of the outer casing or tubing to be detected. Measurements of the inner casing metal radius are valid in the presence of most kinds of scale. Its 21/8-in. diameter affords access through tight restrictions. The tool can operate in gas or liquid environments.

Forewarned Is ForearmedIn 2011, using the EM Pipe Scanner tool, Saudi Aramco conducted a well-casing corrosion moni-toring campaign in a field containing both onshore and offshore wells. Initial scans of seven onshore wells indicated relatively little metal loss and confirmed that the existing ICP system was working satisfactorily. Because of the lack of a sufficiently large power supply, the offshore wells had limited ICP, which raised the possibil-ity of higher corrosion rates.

The EM Pipe Scanner tool was deployed to determine the extent of metal loss from well cas-ings in the offshore portion of the field and to help the operator geographically map wells exhibiting the most severe metal loss. In one campaign, in four adjacent wells that were origi-nally completed in 1976, Saudi Aramco checked to determine whether any of these well had con-centric casings that might soon leak.19 If engi-neers observed metal loss, they planned to analyze the loss profile for the purpose of map-ping and anticipating the likelihood of casing cor-rosion in nearby, nonlogged wells.

The EM Pipe Scanner logs showed varying degrees of metal loss in each of the four subject wells, although the logs indicated a distinct depth correlation among them. One noticeable correla-tion occurred between 2,500 and 2,800 ft [760 and 850 m], where the four wells had casing metal losses ranging from 62% to 65% (left). The operator concluded that other wells in this geographic vicinity were susceptible to significant metal loss and at risk of casing leaks in this depth interval. This conclusion may guide completion decisions for future wells drilled in the area, which could include landing the outermost casing string—

> EM Pipe Scanner logs. The logs for four Saudi Aramco wells showed varying degrees of metal loss (red), remaining thickness (gray) and total measured thickness (green) with respect to depth. A distinct pattern correlation, as well as a similar decrease in total thickness with depth, existed among the wellbores. All wells showed metal losses in the range of 62% to 65% of the outer double casings at a depth of approximately 2,500 ft. The operator used this information to anticipate similar metal loss patterns and expected a comparable level of severity of corrosion in adjacent wells not yet logged.

True

Ver

tical

Dep

th, f

t

True

Ver

tical

Dep

th, f

t

True

Ver

tical

Dep

th, f

t

True

Ver

tical

Dep

th, f

t

1,000

0

Total EM Pipe ScannerThickness

Total EM Pipe ScannerThickness

Total EM Pipe ScannerThickness

Total EM Pipe ScannerThickness

1.5in. 0 1.5in. 0 1.5in. 0 1.5in.

2,000

3,000

4,000

1,000

2,000

3,000

4,000

5,000

1,000

2,000

3,000

4,000

5,000

1,000

2,000

3,000

4,000

5,000

9 5/8-in. shoe9 5/8-in. shoe

9 5/8-in. shoe

9 5/8-in. shoe

65% metal loss ofdouble casings

63% metal loss ofdouble casings

65% metal loss ofdouble casings62% metal loss of

double casings

Autumn 2013 25

typically 133/8-in. casing—deeper than in the previ-ous wells. The original landing depth of 700 ft [213 m] could be extended to a depth of 3,000 ft [914 m] to provide an additional layer of corrosion protection to the inner string. Another solution could be to add a further level of protection by run-ning chrome alloy or coated 133/8-in. casing from 1,000 ft [300 m] to 3,000 ft.

The metal loss profiles from these wells also may influence the operator’s decision to imple-ment more cost-effective and efficient workovers for repairing leaks. For example, the operator could reduce workover costs by running a cement squeeze limited to the depth of significant metal loss rather than incurring additional costs of a liner, casing patch or scab liner, which might be normally recommended if massive metal loss cov-ered a long interval.20

In addition to the acoustic and electromag-netic monitoring techniques discussed, mechani-cal methods are also helpful. A multifinger mechanical caliper tool uses a fundamentally dif-ferent approach. Caliper tools rely on direct physical contact with the pipe wall to make mea-surements and to detect small changes in the tubular wall such as deformations arising from the buildup of scale or metal losses from corro-sion. While they are well established for evaluat-ing internal problems, caliper tools provide no data regarding the condition of the external wall.

The Schlumberger PipeView multifinger cal-iper tool for PS Platform toolstring has been deployed to investigate corrosion in many types of wells but particularly in those with excessive scale and corrosion in which acoustic-based tools cannot be run. The tool can be deployed with 24, 40 or 60 fingers and used in casing diameters ranging from 13/4 in. to 14 in. It pro-vides a mechanical image of the internal tubu-lar corrosion using 3D analysis and visualization software (right).

Measurements over TimeAbu Dhabi Company for Onshore Oil Operations (ADCO) deployed the PipeView tool to measure corrosion over time in a well within a mature and prolific field. The well was originally drilled in 1969 and has been worked over many times. During the most recent workover in 2006, a 7-in.

19. Because the production fluids in these wells were known to be noncorrosive and the tubing-casing annulus contained diesel and corrosion inhibitor, any measured metal loss was assumed to be external only.

20. A cement squeeze is a remedial operation designed to force cement into leak paths in wellbore tubulars and casing strings. Squeeze cementing operations are performed to repair poor primary cement jobs, isolate perforations or repair damaged casings or liners.

>Multifinger caliper tools. Multifinger caliper tools measure the internal diameter in casings. Numerous calipers, or fingers, pressed against the wall of the pipe detect small changes in the pipe inner diameter that may be interpreted as wear or corrosion. In general, multifinger calipers come with varying numbers of fingers; a higher number of fingers is required for larger internal pipe diameters. The PipeView PS Platform new-generation production services multifinger imaging tool, the PMIT-24 fingers tool (left), requires mechanical centralizers (not shown). The PipeView PMIT-40 fingers tool (right) incorporates motorized centralizers. A third version, not shown, has 60 fingers.

PMIT-24 Fingers PMIT-40 Fingers

Motorized centralizers

Motorized centralizers

Calipers

26 Oilfield Review

tieback liner was run and cemented to the sur-face to cover a corroded section of 95/8-in. casing. The operator then drilled a single 57/8-in. horizon-tal well into a previously bypassed carbonate formation. This lateral was completed as a gas lift oil producer.21

Company engineers used naturally produced gas with no corrosion inhibitor treatment as the injection gas, which entered the system through a gas lift side pocket mandrel. Concerned with the corrosion potential posed by the injection gas, ADCO engineers elected to run time-lapse moni-toring surveys with the multifinger imaging tool to identify, quantify and track the growth of internal corrosion in the tubing and estimate a corrosion rate and time-to-failure. ADCO conducted surveys over a three-year period—2009 to 2011—using a 111/16-in., 24-finger version of the tool.

The caliper logs revealed various degrees of corrosion in two sections of the tubing string, one below and one above the injection gas entry point at the side pocket mandrel (above). The lower section, from the bottom of the tubing up to the gas injection point, had experienced a significant degree of corrosion and subsequent metal loss that increased between 2009 and 2011. The upper interval, from the gas lift mandrel to the top of the tubing string, underwent minimal corrosion over the same period and retained its original manufacturing dimensions.

The operator postulated that the injection gas, which enters the produced oil-water flow at the mandrel and flows upward, provides an inhib-itive effect on the production fluids. This effect reduces the corrosion rate in the upper interval, but because the produced fluids below the man-

drel did not contain lift gas, that section experi-enced a higher corrosion rate.

ADCO engineers are still speculating about the exact inhibitor mechanism; one plausible theory holds that the injected gas adds turbu-lence to the production flow and alters the flow regime, which reduces water holdup and water contact with the tubing’s internal surface. This same phenomenon of less corrosion above the gas injection point has been observed in other gas lift wells in which caliper surveys were acquired. A caliper log in a similar well, in combination with a FloView holdup measurement, corroborates the theory that gas injection may be reducing water contact with the tubing (next page). The opera-tor plans to use these results to refine the design of future gas lift well completions to take advan-tage of this effect.

> Corrosion logs obtained from a caliper tool. PipeView service data (top left) and average metal loss versus depth (top right) were recorded in 2009, 2010 and 2011 for the tubing across the point of gas injection. The logs in Track 1 (top and bottom left) include a measurement for nominal internal radius (dashed black line), nominal outer radius (dashed green line), eccentricity (dashed red line), minimum internal radius (solid blue line), maximum internal radius (solid red line) and average internal radius (solid black line) above and below the point of gas injection (top and bottom left, respectively). Track 2 is a trace of each caliper. Track 3 is an image log of thickness loss in the casing. Dark blue indicates the presence of scale, blue to white indicates 0% to 20% thickness loss, white to pink is 20% to 40% loss and orange to red indicates 40% to 80% loss. Pure red (not shown) would indicate 100% loss and a hole in the casing. The average metal loss above the point of gas injection (top right) did not change significantly during the three years, suggesting that the presence of the gas had a mitigating effect on corrosion. A similar plot for the tubing below the point of gas injection (bottom right) showed greater metal loss, which increased over the three-year period, suggesting more-aggressive corrosion.

Dept

h, ft

Dept

h, ft

Y,050

Y,000

X,500

X,550 Aver

age

met

al lo

ss, %

Depth

Average metal loss, 2009Average metal loss, 2010Average metal loss, 2011

Average metal loss, 2009Average metal loss, 2010Average metal loss, 2011

Aver

age

met

al lo

ss, %

50

40

30

20

10

0

50

40

30

20

10

0

1,000 ft

500 ft Depth

Autumn 2013 27

Combining Measurements for Improved Corrosion MonitoringOperators may increase their understanding of the location and extent of downhole tubular cor-rosion by combining information from multiple tools. Kuwait Oil Company (KOC) did this for a well in an onshore field that includes wells that have been producing for more than 60 years.

> Changes to the water holdup profile. A caliper log run in combination with the FloView water holdup probes in an ADCO well shows increasing corrosion over time (Track 2) below the point of gas injection and very little corrosion above the gas injection point. This phenomenon is attributed to a decrease in water holdup above the gas entry point. Analysts believe there is increased gas within the flow regime (right, red dots), which also includes significant water (blue) and oil (green dots). Water holdup, corrected for flowline volume (Track 4), is reduced in the upper section; the corrosion rate is less in the upper section than in the lower section, in which less gas is present. Water holdup is imaged (Track 3); blue represents water and red represents oil and gas.

Gasinjection

Water Holdup,Flow Volume

CorrectedImage View

Average Metal Loss, 2009Dept

h, ft

D,500

E,500

F,500

G,500

H,500

I,500

E,000

F,000

G,000

H,000

I,000

J,000

Average Metal Loss, 2010

Average Metal Loss, 2011

Gamma Ray,2009

Gamma Ray,2010

Gamma Ray,2011

Above theside pocket

mandrel,gas breaks thewater/metal

contact.

Below theside pocket

mandrel, thewater/metal

contact isstable.

Flow regime

Cross-sectionalholdup

distribution00 100%

0 100%

0

0 gAPI

0 gAPI

0 gAPI 100

100

100

100 %

0.9 0.4 1

Several factors, including the age, increased commingling of formation water production and the high CO2 and H2S content of the produced fluids, prompted KOC to examine the corrosion potential of these wells.

During a workover designed to perform a cement squeeze on existing perforations and recomplete the well—which had been drilled and

completed in September 2001 as a single pro-ducer—engineers discovered a leak in the well-bore. To locate the leak zone by quantifying the

21. Gas lift is an artificial lift completion method in which gas is injected into the production tubing to reduce the hydrostatic pressure of the fluid column in the well and the bottomhole pressure. This method allows reservoir liquids to enter the wellbore at a higher flow rate.

28 Oilfield Review

metal loss on the 31/2-in. tubing and the 7-in. and 95/8-in. casing strings, the operator considered pulling the tubing out of the hole and performing pressure tests on the casing. However, this would have added significant cost and taken the well off-line for several weeks.

KOC engineers decided instead to evaluate the integrity of the tubing and casing strings using the PipeView and EM Pipe Scanner tools. The log-ging plan consisted of running the PipeView cali-per tool to assess the metal loss of the tubing and using the EM Pipe Scanner tool to measure the total thickness of the casing strings. By knowing the total combined thickness of the tubing and

casing strings at the outset and subtracting the metal loss from the tubing, engineers were able to attribute any metal loss to the casing strings.

The logging was divided into three sections according to casing design: The first section con-sisted of 31/2-in. and 95/8-in. casings; the second section of 31/2-in., 7-in. and 95/8-in. casings; and the third section consisted of a 7-in. casing. The cali-per logs showed tubing damage in the interval of the well with 31/2-in. and 95/8-in. casings, indicat-ing the presence of holes (above). Also in the first section, the EM Pipe Scanner average thickness measurement revealed metal loss in the outer string of the 95/8-in. casing.

Based on these findings, KOC pulled the tub-ing to confirm the damage. The processed caliper log and a photograph of the damage from the tub-ing show a direct correlation between the corro-sion measurements and the location of the damage (next page, left). The results of this sur-vey gave KOC confidence that it could accurately measure corrosion and identify a leaking interval behind the tubing in wells in the future without having to pull the tubing out of the hole.

Qatar Petroleum also implemented a com-bined corrosion measurement strategy in a well in an offshore field. The well, which was drilled in

> Side-by-side comparison. Logs from the PipeView multifinger caliper tool and the EM Pipe Scanner casing inspection tool run through the interval that contained 31/2-in. and 95/8-in. casing strings showed damage and holes in the 31/2-in. tubing and showed metal loss (Tracks 4, 5 and 6, green shading) on the 95/8-in. casing, including one section that suffered 100% metal loss.

Total Metal Loss

31/2-in. Tubing Thickness

Dept

h, ft

X,000

–8 V 0 360degree–0.8 0.8 0.254 in. 0.395 in.

Casing CollarLocator Depth

Well Schematic

Radii Minus Average

Normal Tubing Thickness

Nominal Total ThicknessTubing Pseudothickness

Double Coil BLong-Spacing Phase

Tubing Plus CasingThickness

Baseline of Outer Casing Thickness

Outer Casing Thickness

Nominal Internal Radius

Nominal Outer Radius

Maximum Internal Radius

MinimumInternal Radius

Average Internal Radius

31/2-in. Tubing Metal Loss 95/8-in. Casing Metal Loss

EM Pipe Scanner Casing Inspection ToolPipeView Multifinger Caliper Tool

Y,000

Z,000

1 1.4 1.9in.

1.4 0.2 in.

in.

0 10.4

0.649 in.

0 in. 9

1.9in. in.

1.4 1.9in.

1.4 1.9in.

1.4 1.9in.

Section: 31/2 in. and 9 5/8 in.

100% metal loss

Autumn 2013 29

> Processed caliper log. The field logs from the PipeView multifinger caliper tool of the interval with 31/2-in. and 95/8-in. casing (top) correlated precisely with the visual damage observed in the retrieved tubing (bottom right). The caliper log (top left) includes measurements for eccentricity (dashed red line), average internal radius (solid black line), maximum internal radius (solid red line), minimum internal radius (solid blue line), excentralization (dashed black line) and nominal outer radius (dashed green line). The caliper log (top center) is composed of three traces that indicate casing collars used for depth correlation (horizontal red line). The image log (top right) in the casing indicates thickness losses. Dark blue indicates scale, blue to white indicates 0% to 20% metal loss, white to pink is 20% to 40% loss and orange to red indicates 40% to 80% loss. Red (circled) indicates 100% loss and a hole in the casing. A 3D processing image (bottom left) based on multifinger caliper data also indicates strong correlation with the damage observed in the retrieved tubing, as do the processed logs (bottom center).

31/2-in. Tubing Thickness

Collar

Multifinger Caliper DisplayNominal Outer Radius

Excentralization

Minimum Internal Radius

Maximum Internal Radius

Average Internal Radius

0 degree 3602 in. 3

1.4 –0.16 0.44

–0.08 0.08

in. in.

in.

1.9

2 in. 3

2 in. 3

2 in. 3

0 in. 1

Average Internal Radius Radii Minus Average

Internal Radius Minus Average PMIT-24

Nominal Outer Radius

Minimum Internal Radius

Maximum Internal Radius

Nominal Internal Radius

Eccentricity

in. 10

Radii Minus Average

Relative Bearing

Holes visible in 3D image and log

1.500Caliper, in.

1.625 1.750

> Casing program. The subject well in a field offshore Qatar contained 31/2-in. tubing, a 7-in. liner and concentric strings of 95/8-in., 133/8-in. and 20-in. casing.

9 5/8-in. casing shoe

3 1/2-in. tubing

13 3/8-in. casing shoe

7-in. liner shoe

20-in. casing shoe

1998, contained 95/8-in., 133/8-in. and 20-in. casing strings, a 7-in. liner and a 31/2-in. production tub-ing string (above). In 2011, the operator observed that the 133/8-in. casing had subsided at the well-head. A pressure test designed to check the integrity of each casing string demonstrated fluid flow in the annular space between the 95/8-in. and 133/8-in. strings and in the annular space separat-ing the 133/8-in. and 20-in. strings. This indicated a leak in the 133/8-in. casing string.

Qatar Petroleum engineers implemented a workover operation, which they began by evaluat-ing the integrity of the cement and presence of corrosion for the 7-in. liner and 95/8-in. casing. An ultrasonic inspection test identified the top of the cement behind the 95/8-in. casing and con-firmed that the 7-in. liner and 95/8-in. casing were free from any significant corrosion or a hole in the casings that might allow fluids communica-tion. Based on the location of the top of the cement, which was identified by the USI tool cement log, Qatar Petroleum engineers were able to determine the interval to cut for casing

30 Oilfield Review

retrieval (above). They could then directly evalu-ate the 133/8-in. casing for corrosion defects.

Engineers next deployed the EM Pipe Scanner tool to evaluate the external casing strings. Despite the fact that measurements were made outside of the recommended specifications, the tool identified an anomaly at a depth above the seabed; the amplitude level across the anomaly was high, and the phase level was low, both of which suggested that less metal was present at the anomaly than would be expected under nor-mal circumstances. This information reinforced the interpretation of the annular pressure test

data and pinpointed the precise location of the hole in the 133/8-in. casing. A PipeView multi-finger imaging tool log was then run to evaluate the 133/8-in. casing; the log showed that the casing was corroded and completely parted at the same depth where the EM Pipe Scanner tool had detected the metal loss (next page). These mea-surements provided Qatar Petroleum with a clear understanding of the location and extent of the corrosion damage such that company engineers could plan a strategy to retrieve the 133/8-in. cas-ing and perform a casing patch operation.

Qatar Petroleum had performed several work-overs on another offshore well in the field and is using the well as a dump flooder, in which pro-duced water is injected into another formation. Because the injected water is untreated, the pro-duction casing regularly experiences corrosion.

The well was originally cased with three sets of steel casing: a 20-in. surface casing, a 133/8-in. intermediate casing and a 95/8-in. production cas-ing. After corrosion problems were detected in 2002, engineers overlapped the production cas-ing with 7-in. casing. The well is perforated in one formation from 6,290 to 6,320 ft [1,918 to 1,926 m]

> Top of cement. Engineers used logs from the USI tool to accurately locate the top of the cement behind the 95/8-in. casing (Tracks 10 and 11); standard USI tool data indicated that the 7-in. liner and the 95/8-in. casing were free of significant corrosion.

–20 –0.08 0.08 –0.08 0.0820in. 0

5 3.5

100dB

0 100dB

0 100dBRPM

0 0.5

6 8

0 degree 0.5

in. 4.2 4.7in.

4.2 4.7in.

4.2 4.7in.

in.

5 3.5in.

5 3.5in.

5 3.5in.0 –6 0.51,000ft/h dB

Casing CollarLocator, Ultrasonic

Average External Radius

Average External Radius

Amplitude ofEccentricity

Wave MinimumAmplitude

Wave AverageAmplitude

Motor SpeedWave Maximum

AmplitudeMaximum Internal

Radius

Maximum InternalRadius

Maximum InternalRadius

Average InternalRadius

Average InternalRadius

Average InternalRadius

Minimum InternalRadius

Minimum InternalRadius

Minimum InternalRadius

Wave AmplitudeMinus Maximum

Cable Speed

Ultrasonic Azimuth

300

400

410

420

430

440

310

320

330

340

350

360

370

380

390

Water

Bonded

Liquid

Microdebonding

Annulus

Water

Annulus

Dept

h, ft

Casing Casing

5 3.5in.

5 3.5in. 0.1 0.6 0 01 –2.0 8.08.0Mrayl Mraylin.

0.1 0.6in.

0.1 0.6in.

5 3.5in.

5 3.5in.

Maximum Thickness

Average Thickness

Cement AcousticImpedance

Average Casing Thickness

Internal RadiiMinus Average Microdebonding

Minimum Thickness 01

Microdebonding Ratio

01

Cement Measurements/Total

GasMeasurements/Total

Top of cementNo severe metal

loss detected

Autumn 2013 31

and in another from 6,523 to 7,030 ft [1,988 to 2,143 m]. Produced water from both formations is injected into a formation from 7,492 to 7,690 ft [2,284 to 2,344 m].

As part of regular operational monitoring and assessment of the well, Qatar Petroleum engi-neers deployed the EM Pipe Scanner tool to eval-uate the well for corrosion. The tool’s findings indicated significant metal loss across the 7-in. and the 95/8-in. sections. At approximately 6,250 ft [1,900 m], the tool indicated a thickness of about 0.28 in. [0.71 cm], much less than the expected nominal thickness of 0.797 in. [2.03 cm], which implied a metal loss of approximately 65%. The well’s history and the operator’s local experience in the field suggested that the entire 95/8-in. cas-ing was completely corroded and the 7-in. casing was partially corroded with about 10% metal loss. The EM Pipe Scanner tool’s high-frequency image measurement confirmed that the 7-in. inner casing was not fully penetrated by corro-sion, which indicated that the inside wall of the pipe was in good condition.

Improved Corrosion Mitigation Through Management Downhole corrosion monitoring tools help engi-neers understand the physical condition of tub-ing and casing strings. Operators can then make more-informed and cost-effective mitigation and repair decisions. But as companies continue to search for more streamlined and holistic meth-ods for protecting their assets and extending the producing life of their wells, service providers have worked to improve monitoring capabilities.

For example, the advent of online, near real-time measurement capabilities has brought about a natural progression to developing corro-sion monitoring workflows and software plat-forms that maximize the usefulness of recorded data. These platforms use advances in informa-tion and communication technology to improve oil and gas E&P efforts with the objectives of opti-mizing field operations and avoiding nonproduc-tive time.

Schlumberger engineers are working to inte-grate corrosion measurement data gathering into overall field operations. These efforts are focused in three software-based management platforms. The Petrel E&P software platform provides oper-ators and service companies with a reservoir-level view of field optimization by allowing asset teams to build collaborative workflows based on geomechanical, geochemical and reservoir fluid properties. Along with information such as the reservoir temperature and pressure and the expected corrosive characteristics of production

fluids in the reservoir, the Petrel software helps guide well planners in making well decisions to ensure a high-integrity wellbore constructed of appropriate alloys.

The Techlog wellbore software platform fur-ther advances this evaluation by providing well-bore-centered workflows to identify corrosion risks. These workflows incorporate fluid compo-sition and flow rate data to flag any potential corrosion-induced wellbore problems, allowing the operator to make construction and comple-tion decisions that minimize corrosion’s impact. The Avocet production operations software plat-form combines well operations and production data management systems to deliver a clear and comprehensive picture of operations at the sur-face. The Avocet software accepts corrosion

data recorded from various monitoring tech-niques and analyzes such data for their impact on production. The software flags those areas with higher corrosion rates or a history of corro-sion-related events, and as a result, the operator can prioritize corrosion mitigation efforts and implement suitable preventive measures.

As the industry moves into more-aggressive corrosion environments and technically demand-ing production regions, corrosion monitoring advances such as these must continue to expand and evolve if operators are to remain both profit-able and environmentally responsible. — TM

> Significant metal loss. Even though the EM Pipe Scanner tool was run outside of its specified range for amplitude and phase, it detected significant metal loss across the three casing strings (dashed red box, top left). The 3D (top center) and 2D (top right) visualizations produced from the PipeView multifinger caliper tool log run in the 133/8-in. casing indicate the casing had corroded to the point that it had parted (bottom left and right) at the depth where the EM Pipe Scanner tool had detected metal loss.

Metal Loss

20-in. casing

13 3/8-in. casing

9 5/8-in. casing

90

Depth, ft –60 dB

Double Coil BLong-Spacing Phase

Double Coil BLong-Spacing Amplitude

0

40 degree 400

90

86

82

78

74

94

98

102

106

110

114

118

122

Dept

h, ft

6.688

6.578

6.469

6.359

6.250

Calip

er, i

n.

6.6886.469 6.5786.250 6.359

Caliper, in.

100

32 Oilfield Review

Geomagnetic Referencing—The Real-Time Compass for Directional Drillers

To pinpoint the location and direction of a wellbore, directional drillers rely on

measurements from accelerometers, magnetometers and gyroscopes. In the past,

high-accuracy guidance methods required a halt in drilling to obtain directional

measurements. Advances in geomagnetic referencing now allow companies to use

real-time data acquired during drilling to accurately position horizontal wells,

decrease well spacing and drill multiple wells from limited surface locations.

Andrew BuchananEni US Operating Company Inc.Anchorage, Alaska, USA

Carol A. Finn Jeffrey J. LoveE. William Worthington US Geological SurveyDenver, Colorado, USA

Fraser LawsonTullow Ghana Ltd.Accra, Ghana

Stefan MausMagnetic Variation Services LLCBoulder, Colorado

Shola OkewunmiChevron CorporationHouston, Texas, USA

Benny PoedjonoSugar Land, Texas

Oilfield Review Autumn 2013: 25, no. 3. Copyright © 2013 Schlumberger.For help in preparation of this article, thanks to Essam Adly, Muscat, Oman; Goke Akinniranye, The Woodlands, Texas; James Ashbaugh and Robert Kuntz, Pennsylvania General Energy Company, LLC, Warren, Pennsylvania, USA; Nathan Beck, Anchorage; Luca Borri, Jason Brink and Joseph Longo, Eni US Operating Co. Inc., Anchorage; Daniel Cardozo, St. John’s, Newfoundland, Canada; Pete Clark, Chevron Energy Technology Company, Houston; Steve Crozier, Tullow Ghana Ltd., Accra, Ghana; Mike Hollis, Chesapeake Energy, Oklahoma City, Oklahoma, USA; Christopher Jamerson, Apache Corporation, Tulsa; Xiong Li, CGG GravMag Solutions, Houston; Ross Lowdon, Aberdeen; Diana Montenegro Cuellar, Bogotá, Colombia; Ismail Bolaji Olalere, Shell Nigeria, Port Harcourt, Nigeria; Irina Shevchenko, Michael Terpening and John Zabaldano, Houston; Tim White, US Geological Survey, Denver; and the Government of Newfoundland and Labrador, Department of Natural Resources, St. John’s, Newfoundland, Canada.PowerDrive is a mark of Schlumberger.

For a variety of reasons, operating companies need to know where their wells are as they are being drilled. Many of today’s deviated and hori-zontal wells no longer simply penetrate a reser-voir zone but must navigate through it laterally to contact as much of the reservoir as possible. Precise positioning of well trajectories is required to optimize hydrocarbon recovery, determine where each well is relative to the reservoir and avoid collisions with other wells.

To accomplish these objectives, drillers require directional accuracy to within a fraction of a degree. To achieve this level of accuracy, they use measure-ment-while-drilling (MWD) tools that include accel-erometers and magnetometers that detect the Earth’s gravitational and magnetic fields; they also use sophisticated procedures to compensate for measurement perturbation. As drillers have found success with these tools and become more depen-dent on them for well guidance, the need for accu-rately quantified positional uncertainty that takes into account all measurement error has also increased. For some applications, the uncertainty is as important as the position itself.

This article reviews aspects of wellbore sur-veying, focusing on modern techniques for mag-netic surveying with MWD tools. To understand the operation of and uncertainty associated with magnetic tools, we examine important aspects of the Earth’s magnetic field and its measurement. Examples from the USA, Canada, offshore Brazil and offshore Ghana illustrate the application of new techniques that improve measurement accu-racy and thus effect considerable reduction in magnetic tool survey error.

1. Borehole orientation may be described in terms of inclination and azimuth. Inclination refers to the vertical angle measured from the down direction—the down, horizontal and up directions have inclinations of 0°, 90° and 180°, respectively. Azimuth refers to the horizontal angle measured clockwise from true north—the north, east, south and west directions have azimuths of 0°, 90°, 180° and 270°, respectively. For more on borehole orientation: Jamieson AL: Introduction to Wellbore Positioning. Inverness, Scotland: University of the Highlands and Islands, 2012, http://www.uhi.ac.uk/en/research-enterprise/wellbore-positioning-download (accessed June 18, 2013).

2. Griswold EH: “Acid Bottle Method of Subsurface Well Survey and Its Application,” Transactions of the AIME 82, no. 1 (December 1929): 41–49.

>Mechanical drift indicator. This downhole device measures drift, or deviation from vertical, using a pendulum, or the “plumb bob,” principle. The sharp-tipped pendulum is lowered onto a disk into which it punches two holes that mark an initial measurement then a verification measurement. In this example, the inclination is 3.5°. The technique gives no indication of azimuth but may be reliable for surface hole intervals and shallow vertical wells in which dogleg severity and inclination are not significant. [Adapted from Gatlin C: Petroleum Engineering Drilling and Well Completions. Englewood Cliffs, New Jersey, USA: Prentice-Hall, Inc. (1960): 143.]

6°

4°

2°

Plumb bob

Disk

Clock

Punch marks show3.5° inclination

Drift indicator disk

Autumn 2013 3333

Historical PerspectiveTraditionally, wellbores were drilled vertically and were widely spaced. Well spacing decreased as fields matured, regulations tightened and res-ervoirs were targeted in remote areas. Over time, drilling multiple horizontal wells from a single pad became common practice. Today, more than a dozen wells may fan out into the reservoir from a single offshore platform or onshore drilling pad.

Pad drilling—grouping wellheads together at one surface location—necessitates fewer rig moves, requires less surface area disturbance and

makes it easier and less expensive to complete wells and produce hydrocarbons. However, the introduction of horizontal drilling and closer well-bore spacing has intensified the need for accurate wellbore positioning and for processes to prevent collisions between the bit and nearby wellbores.

Before the introduction of modern steerable downhole motors and advanced tools to measure hole inclination and azimuth, directional or hori-zontal drilling was much slower than vertical drilling because of the need to stop regularly and take time-consuming downhole surveys. The

directional driller stopped drilling to measure wellbore inclination and azimuth.1

The oldest survey method entailed lowering a glass bottle of acid downhole and holding it station-ary long enough for the acid to etch a horizontal ring in the bottle. The ring’s position was interpreted for inclination once the device was retrieved.2

Another simple survey tool is the single shot mechanical drift indicator (previous page). Magnetic single shot (MSS) and multishot (MMS) surveys have also been used to record inclination and magnetic azimuth. For those surveys, the tool

34 Oilfield Review

took photographs, or shots, of compass cards downhole while the pipe was stationary in the slips. Photographs were taken every 27 m [90 ft] during active changes of angle or direction and every 60 to 90 m [200 to 300 ft] while drilling straight ahead. The introduction of downhole mud motors in the 1970s and the development of rug-gedized sensors and mud pulse telemetry of MWD data enabled the use of continuously updated digi-tal measurements for near real-time trajectory adjustments. Most wells are now drilled using sur-vey measurements from modern MWD tools.

Well Survey BasicsToday, directional drillers rely primarily on real-time MWD measurements of gravitational and magnetic fields using ruggedized triaxial acceler-ometers and magnetometers. Other categories of survey tools include magnetic multishot tools, inclination-only tools and a family of tools based on the use of gyroscopes, or gyros.3 Unlike MWD tools, many of these specialty tools are run as wireline services, thus requiring cessation of the drilling process. Increasingly, however, gyro-scopic tools are also being incorporated into downhole steering and surveying instruments for use while drilling.

Triaxial accelerometers measure the local gravity field along three orthogonal axes. These measurements provide the inclination of the tool axis along the wellbore as well as the toolface relative to the high side of the tool.4 Similarly, tri-axial magnetometers measure the strength of the Earth’s magnetic field along three orthogonal axes. From these measurements and the acceler-ometer measurements, the tool determines azi-muthal orientation of the tool axis relative to magnetic north. Conversion of magnetic mea-surements to geographic orientation is at the heart of MWD wellbore surveying. The key mea-surements are magnetic dip (also called mag-netic inclination), total magnetic field and magnetic declination (above).5

A variety of tools exploit gyroscopic princi-ples. These systems are unaffected by ferromag-netic materials, giving them an advantage over magnetic tools in some drilling scenarios. Some tools take measurements at discrete intervals of measured depth (MD) along the well path when the survey tool is stationary; others operate in a continuous measurement mode. North-seeking gyrocompasses (NSGs) make use of gyroscopes and the rotation of the Earth to automatically find geographic north. Rate gyros provide an out-put proportional to the turning rate of the instru-ment and may be used to determine orientation

as the survey tool continuously traverses the well path. Surveying engineers also use them in gyro-compassing mode, in which the stationary tool responds to the horizontal component of the Earth’s rotation rate. The use of rate gyros has reduced errors—such as geographic reference errors and unaccountable measurement drift—that are associated with conventional gyros. Unfortunately, because they are taken while the tool is stationary, gyro surveys carry operational risk and rig time cost associated with wellbore conditioning when drilling is stopped.6

In some intervals, significant magnetic inter-ference from offset wellbores makes accurate magnetic surveying impossible. To address this limitation, scientists developed gyro-while- drilling methods. Tool design engineers are extending the operational limits of some com-mercial gyro-while-drilling survey systems to the full range of wellbore inclinations.

For some situations, surveying engineers combine gyroscopic and magnetic surveying. One of the combined techniques—inhole referenc-ing—makes use of highly accurate gyroscope measurements in shallow sections to align subse-quent data obtained using magnetic surveys in deeper sections.7 In highly deviated and extended-reach wells, this approach delivers lev-els of accuracy comparable to those acquired with wireline gyroscopic surveys without incur-ring the added time and costs. In these inhole referencing systems, gyroscopic measurements are used in shallow near-vertical wellbore sec-tions in the vicinity of casing until MWD mag-netic surveys can be obtained free of interference and in longer-reach sections in which inclina-tions increase. An additional benefit of using both gyro and MWD surveys is the detection of gross error sources in either tool.

Positional UncertaintyDrillers use positional uncertainty estimates to determine the probability of striking a geologic target and of intersecting other wellbores.8 They base the estimates on tool error model predic-tions, which themselves depend on quality con-trol (QC) of survey data. Survey tool quality checks help identify sources of error, often with redundant surveys as independent cross-checks.

For most survey tools, the outputs are azi-muth, inclination and measured depth. Errors in each measurement may occur because of both the tool and the environment. Accuracies avail-able from stationary measurements made with standard MWD tools are on the order of ±0.1° for inclination, ±0.5° for azimuth and ±1.0° for toolface.

>Magnetic field orientation. At any point P, the magnetic field vector (red) is commonly described in terms of its direction, its total magnitude, F, in that direction and H and Z, the local horizontal and vertical components of F. The angles D and I describe the orientation of the magnetic field vector. The declination, D, is the angle in the horizontal plane between H and geographic north. The inclination, I, is the angle between the magnetic field vector and the horizontal plane containing H. Of these measurements, D and I are required to convert the compass orientation of a wellbore to its geographic orientation. The absolute magnitudes of F, Z or H are used for quality control and calibration.

East

West

Down

Magnetic field vector

Magnetic north

Geographic north

Z

P

X

HY

F

D

I

Autumn 2013 35

A surveying engineer’s ability to determine borehole trajectory depends on the accumulation of errors from wellhead to total depth. Rather than specifying a point in space, surveying engi-neers consider wellbore position to be within an ellipsoid of uncertainty (EOU). Typically, the uncertainty in the lateral direction is larger than in the vertical or along-hole directions. When dis-played continuously along the wellbore, the EOU presents a volume shaped like a flattened cone surrounding the estimated borehole trajectory (right). The combined effects of accumulated error may reach values of 1% of measured well depth, which could be unacceptably large for long wellbores.9

The Industry Steering Committee for Wellbore Survey Accuracy, ISCWSA—now the SPE Wellbore Positioning Technical Section, WPTS—has promoted development of a rigorous mathe-matical procedure for combining various error sources into one 3D uncertainty ellipse.10 External effects on accuracy include axial mis-alignment, BHA deflection, unmodeled geomag-netic field variations and drillstring-induced interference. The latter two factors dominate the performance of magnetic tools and their error models; such models depend on the resolution of the geomagnetic reference model in use.11

The Geomagnetic Field To make use of magnetic measurements for find-ing direction, it is necessary to take into account the complexity of the geomagnetic field. The geo-magnetic field surrounds the Earth and extends into nearby space.12 The total magnetic field mea-sured near the Earth’s surface is the superposition of magnetic fields arising from a number of time-

3. This family includes conventional gyros, rate gyros, north-seeking gyros, mechanical inertial gyros and ring laser inertial gyros. For more on gyros: Jamieson AL: “Understanding Borehole Surveying Accuracy,” Expanded Abstracts, 75th SEG Annual International Meeting and Exposition, Houston (November 6–11, 2005): 2339–2340.

Jamieson, reference 1. 4. Gravity, or high side, toolface is the orientation of the

survey instrument in the borehole relative to up. Magnetic toolface is the orientation of the survey instrument relative to magnetic north, corrected to a chosen reference of either grid north or true north. Most MWD systems switch from a magnetic toolface to a high side toolface once the inclination exceeds a preset threshold typically set between 3° and 8°. For more on instrument orientation: Jamieson, reference 1.

5. By international agreement, magnetic field orientation may be described in terms of dip (also referred to as inclination) and declination. Dip is measured positively downward from the horizontal direction—the down,

> Planned well trajectories showing slices of the ellipsoids of uncertainty (EOUs) obtained from standard MWD (blue) and from higher accuracy MWD (red) surveys. The azimuthal and inclination uncertainties are in the XY plane perpendicular to the borehole. The depth uncertainty is along the Z-axis of the borehole. When shown at a dense series of points along the well trajectory, they form a “cone of uncertainty.” The high-accuracy method delivers a wellbore with smaller positional uncertainty. (Adapted from Poedjono et al, reference 32.)

XY Z

200 ft

1,000 ft1,000 ft

horizontal and up directions have dips (inclinations) of 90°, 0° and –90°, respectively. Declination is defined similarly to hole azimuth. For more on magnetic field orientation: Campbell WH: Introduction to Geomagnetic Fields, 2nd ed. Cambridge, England: Cambridge University Press, 2003.

6. Gyro surveys conducted on wireline in openhole sections carry the risk of stuck survey tools. Surveys made through drillpipe when the drilling is stopped carry the risk of stuck drillpipe. Additionally, operators usually perform a hole conditioning cleanup cycle after drilling is stopped. These combined operations may require many hours of rig time.

7. Thorogood JL and Knott DR: “Surveying Techniques with a Solid-State Magnetic Multishot Device,” SPE Drilling Engineering 5, no. 3 (September 1990): 209–214.

8. Ekseth R, Torkildsen T, Brooks A, Weston J, Nyrnes E, Wilson H and Kovalenko K: “High-Integrity Wellbore Surveying,” SPE Drilling & Completion 25, no. 4 (December 2010): 438–447.

9. For typical well depths and step-out, or horizontal reach, the dimensions of the uncertainty envelope may be on the order of 100 ft [30 m] or more unless action is taken to correct error sources and run high-accuracy surveys. This may exceed the size of the target and increase the risk of unsuccessful wellbore steering. For more on the calculation, extent and causes of positional uncertainty: Jamieson, references 1 and 3.

10. For more on tool error model selection and the accepted industry standard ISCWSA error models for magnetic tools: Williamson HS: “Accuracy Prediction for Directional Measurement While Drilling,” SPE Drilling & Completion 15, no. 4 (December 2000): 221–233.

For more on error models for gyroscopic tools: Torkildsen T, Håvardstein ST, Weston JL and Ekseth R: “Prediction of Wellbore Position Accuracy When Surveyed with Gyroscopic Tools,” paper SPE 90408, presented at the SPE Annual Technical Conference and Exhibition, Houston, September 26–29, 2004.

11. Williamson, reference 10.12. Love JJ: “Magnetic Monitoring of Earth and Space,”

Physics Today 61, no. 2 (February 2008): 31–37.

36 Oilfield Review

varying physical processes that are grouped into four general components: the main magnetic field, the crustal field, the external disturbance field and local magnetic interference.13 The significance of these contributions to direction, strength and stability of the total magnetic field varies with geo-graphic region and with magnetic survey direction. The importance of accounting for each component in the measurement depends on the purpose and required accuracy of the survey.

Physicists have determined that the Earth’s main magnetic field is generated in the Earth’s fluid outer core by a self-exciting dynamo pro-cess. Approximately 95% of the total magnetic field measured at Earth’s surface comes from this main field, a significant portion of which may be described as the field of a dipole placed at the Earth’s center and tilted approximately 11° from the Earth’s rotational axis (left). The magnitude of the main magnetic field is nearly 60,000 nT near the poles and about 30,000 nT near the equator.14 However, there are significant non-dipole contributions to the main magnetic field that complicate its mathematical and graphical representation (below left). As an additional complication, the main field varies slowly

13. Akasofu S-I and Lanzerotti LJ: “The Earth’s Magnetosphere,” Physics Today 28, no. 12 (December 1975): 28–34.

Jacobs JA (ed): Geomagnetism, Volume 1. Orlando, Florida, USA: Academic Press, 1987.

Jacobs JA (ed): Geomagnetism, Volume 3. San Diego, California, USA: Academic Press, 1989.

Merrill RT, McElhinny MW and McFadden PL: The Magnetic Field of the Earth: Paleomagnetism, the Core, and the Deep Mantle. San Diego, California: Academic Press, International Geophysics Series, Volume 63, 1996.

Campbell, reference 5. Lanza R and Meloni A: The Earth’s Magnetism: An

Introduction for Geologists. Berlin: Springer, 2006. Auster H-U: “How to Measure Earth’s Magnetic Field,”

Physics Today 61, no. 2 (February 2008): 76–77. Love, reference 12.14. The symbol B is often used for magnetic induction,

the quantity that is sensed by magnetometers. The SI unit for B is the Tesla (T), and the centimeter-gram-second (cgs) unit is the Gauss (G); the common unit is the gamma, which is 10–9 T = 1 nT.

15. Time variations, called secular variations, necessitate periodic updating of magnetic field maps and models. These variations are caused by two types of processes in the Earth’s core. The first is related to the main dipole field and operates on time scales of hundreds or thousands of years. The second is related to nondipole field variations at time scales on the order of tens of years. For more on secular variations: Lanza and Meloni, reference 13.

16. Remanent magnetism of rocks results from exposure of magnetic materials in the rocks to the Earth’s magnetic field when the rocks were formed. Igneous rocks retain thermoremanent magnetization as they cool. In some rocks, remanent magnetization arises when magnetic grains are formed during chemical reactions. Sedimentary rocks retain remanent magnetization when magnetic grains align with the magnetic field during sediment deposition. Remanent magnetism also occurs in ferromagnetic materials, such as the steel in casing or drillpipe, as a result of exposure to the Earth’s magnetic field or industrial magnetic field sources.

> Simplified geomagnetic field. The Earth’s main geomagnetic field is portrayed as the ideal magnetic field of a geocentric tilted dipole with poles at the core of the Earth (brown shading). Lines of magnetic flux (red) emanate outward through the surface of the Earth near the geographic south pole and reenter near the geographic north pole. Those positions along the axis of the dipole are the magnetic south and north poles, although the polarity of the internal dipole is the opposite. The geographic north and south poles lie on the Earth’s axis of rotation. Both axes are tilted relative to the plane of the Earth’s rotational orbit.

Axis of magnetic poles

Line in orbital plane

Axis of Earth’s rotation

S

N

> Values of declination along lines of equal declination (isogonic lines) of the Earth’s magnetic field. In the areas surrounded by red lines, or the lines of equal positive declination, a compass points to the east of true north. Lines of equal negative declination, for which the compass points to the west of true north, are blue. Along the green, agonic lines, for which declination equals zero, the directions to magnetic north and true north are identical. The field shown is the International Geomagnetic Reference Field for the year 2010. [Adapted from “Historical Main Field Change and Declination,” CIRES Geomagnetism, http://geomag.org/info/declination.html (accessed June 24, 2013).]

10

10

10

10

20

30

–20

–20

–20

–30–40

–10

–10

Autumn 2013 37

because of changes within the Earth’s core. The relative strengths of nondipole components change, and the magnetic dipole axis pole posi-tion itself wanders over time (above).15

The magnetic field associated with the Earth’s crust arises from induced and remanent magnetism.16 The crustal field—also referred to as the anomaly field—varies in direction and strength when measured over the Earth’s sur-face (above right). It is relatively strong in the vicinity of ferrous and magnetic materials, such as in the oceanic crust and near concentrations of metal ores, and is the focus of geophysical mineral exploration.

The disturbance field is an external magnetic field arising from electric currents flowing in the ionosphere and magnetosphere and “mirror-cur-rents” induced in the Earth and oceans by the external magnetic field time variations. The dis-turbance field is associated with diurnal field variations and magnetic storms (see “Blowing in the Solar Wind: Sun Spots, Solar Cycles and Life on Earth,” page 48). This field is affected by solar activity (solar wind), the interplanetary mag-netic field and the Earth’s magnetic field (right).

> Variation of the position of the northern magnetic pole between 1990 and 2010. Magnetic declination (red and blue lines) from the International Geomagnetic Reference Field model is shown for 2010. The green dot represents the position of the magnetic dip pole in 2010; the yellow dot represents the position of that pole in 1990. The agonic lines, for which declination equals zero in 2010, are highlighted in green. If a compass at any location points to the right of true north, declination is positive, or east (red contours), and if it points to the left of true north, declination is negative, or west (blue contours). [Adapted from “Historical Magnetic Declination,” NOAA National Geophysical Data Center, http://maps.ngdc.noaa.gov/viewers/historical_declination/ (accessed June 24, 2013).]

Year 2010

> Geomagnetic crustal field. Airborne measurements of the strength of the magnetic field provide data that are used to determine the anomalous contribution from earth crustal materials. The total intensity anomaly (TIA) is the difference between the magnitude of the total field and that of the main magnetic field. The TIA field over western Canada; Alaska, USA; and the northwest continental US varies from –300 nT (blue) to +400 nT (pink). The mean total field strength is about 55,000 nT in this region. The crustal field shows local intensity ridges, with variation on a much finer spatial scale than that of the main magnetic field. [Adapted from “Magnetic Anomaly Map of North America,” USGS, http://mrdata.usgs.gov/geophysics/aeromag-na.html (accessed July 23, 2013).]

400

150

90

70

50

Tota

l int

ensi

ty a

nom

aly,

nT 30

20

10

0

–10

–20

–30

–40

–60

–80

–125

–175

–300

AlaskaC A N A D A

P a c i f i c O c e a n

> Distortion of the Earth’s magnetosphere from the solar wind. The sun emits a flux of particles, called the solar wind, which consists of electrons, protons, helium [He] nuclei and heavier elements. The Earth’s magnetic field is confined by the low-density plasma of the solar wind and the interplanetary magnetic field (IMF) that accompanies it. These distort the Earth’s magnetic field away from its dipolar shape in the magnetosphere, the extensive region of space bounding the Earth. The field becomes compacted on the sunward side and elongated on the opposite side. The solar wind produces a variety of effects, including the magnetopause, radiation belts and the magnetotail. Time-varying interactions of the magnetosphere with the solar wind produce magnetic storms and the external disturbance field.

Solar wind

Bow shock

Magnetopause

Magnetotail

Earth

Magnetosheath

Magnetic field lines

Van Allen radiation belts

38 Oilfield Review

The external magnetic field exhibits varia-tions on several time scales, which may affect the applicability of magnetic reference models.17 Very long-period variations are related to the solar cycle of about 11 years. Short-term variations arise from daily sunlight variation, atmospheric tides and diurnal conductivity variations. Irregular time variations are influenced by the solar wind. Perturbed magnetic states, called

magnetic storms, arise and show impulsive and unpredictable rapid time variations.

On the local scale, nearby structures such as rigs and wells may induce magnetic interference. Drillstring remanent magnetization and mag-netic permeability contribute to perturbations of the measured magnetic field (above). Operators may use nonmagnetic drill collars to reduce these effects along with software techniques to compensate for them.

Magnetic Field Measurements, Instrumentation and ModelsPhysicists have developed a variety of sophisti-cated instruments for measuring magnetic fields.18 Of particular interest for geomagnetic referencing are the instruments that scientists use within magnetic observatories on the Earth’s surface and those that surveying engineers use in the oil field for downhole MWD surveying.

Proton precession and Overhauser magne-tometers, which measure the Earth’s magnetic field, are based on the phenomenon of nuclear paramagnetism and the tendency of atomic nuclei with a magnetic spin to orient along the dominant magnetic field. During this process, a current-induced magnetic field is applied and removed intermittently, and then the frequency of precession is measured as protons in the sen-sor fluid precess under the influence of the Earth’s magnetic field. The Overhauser magne-tometer makes use of additional free electrons in the sensing fluid and the application of a strong radio frequency polarizing field to enable contin-uous measurement of the precession frequency. The 14 US-based US Geological Survey (USGS) magnetic observatories use Overhauser magne-tometers to provide absolute measurements of magnetic field intensity.19 These magnetometers achieve absolute accuracy on the order of 0.1 nT.

Fluxgate magnetometers operate by driving the cores of magnetic circuits into saturation and measuring slight asymmetries that arise from the additional contribution of the Earth’s magnetic field. These instruments give nonabsolute mag-netic measurements along a particular direction, with resolution as fine as 0.01 nT.20 The instru-ments are used in surface observatories and in ruggedized downhole MWD equipment, although some instruments are temperature sensitive and require stabilization through mechanical design.

Magnetic field models provide values for mag-netic declination, magnetic inclination and total magnetic field at points on the surface of the Earth; scientists use these models to transform magnetic measurements to directions in the geo-graphic coordinate system. Various organizations have developed geomagnetic reference models using global magnetic field measurements taken from satellite, aircraft and ships. These organiza-tions include the US National Oceanic and Atmospheric Administration (NOAA), the NOAA National Geophysical Data Center (NGDC), the British Geological Survey (BGS) and the International Association of Geomagnetism and Aeronomy (IAGA). The models differ in their resolution in space and time (left).

> Contributions to the total observed magnetic field. During periods of solar quiet, the discrepancy between the observed field, Bobserved (red), and the main magnetic field, Bm (green), is largely due to the local crustal field Bc (blue) and the drillstring interference, Bint (yellow). At other times, the external disturbance field also makes a contribution. (Adapted from Poedjono et al, reference 30.)

Bint

Bint

Bc

Bm

Bobserved

>Magnetic field reference models. Several groups and organizations have developed reference models of differing resolution; the models are updated at various intervals. In the Order column, order increases with the complexity of the model and in this case refers to spherical harmonic models. These models construct the global magnetic field as a sum of terms of varying order and degree. Terms of order “n” have a total of n circular nodal lines on the sphere at which the magnetic field contribution is zero. The orientation of the lines depends on the combination of order and degree. Resolution corresponds to the wavelength of the highest order term.

Model Organization Order Resolution, km

WMM

IGRF IAGA

BGGM BGS

EMM and HDGM

NOAA, NGDC and BGS

NOAA and NGDC

Update Interval

5 years

5 years

1 year

5 years and 1 year

12

13

50

720

3,334

3,077

800

56

Autumn 2013 39

The World Magnetic Model (WMM) character-izes the long-wavelength portion of the magnetic field that is generated in the Earth’s core; it does not represent the portions that arise either in the crust and upper mantle or from the disturbance field generated in the ionosphere and magneto-sphere.21 Consequently, magnetic measurements may show discrepancies when referenced to the WMM alone. Local and regional magnetic declina-tion anomalies occasionally exceed 10°, and decli-nation anomalies on the order of 4° are not uncommon but are usually of small spatial extent. To account for secular variation, the WMM is updated every five years. An international task force formed by the IAGA has released International Geomagnetic Reference Field IGRF-11, a series of mathematical models of the Earth’s main magnetic field and its rate of change. These models have reso-lution that is comparable to that of the WMM.22

Directional drilling requires higher resolu-tion models than WMM or IGRF alone. The BGS Global Geomagnetic Model (BGGM), widely used in the drilling industry, provides the main mag-netic field at 800-km [500-mi] resolution and is updated annually.23 The Enhanced Magnetic Model (EMM) improves greatly on this spatial resolution. The EMM and a successor, the High-Definition Geomagnetic Model (HDGM), resolve anomalies down to 56 km [35 mi], an order of magnitude improvement over previous models. By accounting for a larger waveband of the geo-magnetic spectrum, the HDGM improves the accuracy of the reference field, which in turn improves the reliability of wellbore azimuth determination and enables high-accuracy drill-string interference correction.24

Improving Well Position AccuracyTo place wellbores accurately when using mag-netic guidance, surveying engineers must account for or eliminate two important sources of survey error: interference caused by magnetized ele-ments in the drillstring and local variations between magnetic north and true, or geographic, north. Analysis of data from multiple wellbore sur-vey stations, or multistation analysis (MSA), has become the key to addressing drillstring interfer-ence. Surveying engineers use geomagnetic refer-encing, which accounts for the influence of the crustal field and the time-varying disturbance field as well as secular variations in the main mag-netic field.

Multistation analysis—MSA is a technique that helps compensate for drillstring magnetic interference, which can affect downhole mag-netic surveys.25 Drillstring components generate local disturbances to the Earth’s magnetic field because of their magnetic permeability and remanent magnetization. Using tools manufac-tured with nonmagnetic materials to isolate directional sensors from magnetized drillstring components is beneficial, but the use of such tools may be imperfect or impractical because they may impact the cost or performance of the BHA. An alternative is to characterize the magni-tude of the disturbance associated with the BHA so that its influence is predictable.

The MSA technique assesses the magnetic signature of the BHA by comparing the Earth’s main magnetic field with magnetic data acquired at multiple survey stations. The magnitude of the perturbation depends on the orientation of the tool relative to the magnetic field direction. With

sufficient data, the method determines a robust correction of the BHA disturbance to be applied for each particular well orientation.

Multistation analysis is an improvement over the earlier technique of single station analysis in which compensation is estimated and applied to each survey station independently. Now com-monly used in the industry, MSA generally reduces directional uncertainty and aids in pen-etration of smaller reservoir targets than were previously achievable. The technique can elimi-nate some gyrocompass runs, thus reducing oper-ational costs. Service companies have developed data requirements and acceptance criteria that have to be fulfilled when applying MSA, and an industry standard has been proposed.26

Geomagnetic referencing—Another tech-nique for improving wellbore position accuracy, geomagnetic referencing provides the mapping between magnetic north and true north that is necessary to convert magnetically determined orientations to geographic ones on a local scale. The mapping must account for secular variations in the main magnetic field model and include an accurate crustal model. Furthermore, it must incorporate the time-varying disturbance field when it is significant. The Schlumberger geomag-netic referencing method builds a custom model of the geomagnetic field, with all its magnetic field components, to minimize the error in the mapping between magnetic and true north.27

Annually updated magnetic field models such as the BGGM or HDGM accurately track secular variations of the main magnetic field. Surveying engineers employ such models as the foundation for a custom model. They use various techniques

17. During quiet periods of solar activity, daily field variations, called diurnal variations, can have magnitudes of about 20 nT at midlatitudes and up to about 200 nT in equatorial regions. During periods of heightened solar activity, magnetic storms may persist for several hours or several days with deviations in magnetic intensity components on the order of several tens to hundreds of nT at midlatitudes. In auroral regions, the disturbances occasionally reach 1,000 nT, and the declination angle can vary by several degrees or more. For more on magnetic reference models: Lanza and Meloni, reference 13 and Campbell, reference 5.

18. Campbell, reference 5. Lanza and Meloni, reference 13. Auster, reference 13. 19. Love JJ and Finn CA: “The USGS Geomagnetism

Program and Its Role in Space Weather Monitoring,” Space Weather 9, no. 7 (July 2011): S07001-1–S07001-5.

20. Auster, reference 13.21. For more on the World Magnetic Model (WMM):

Maus S, Macmillan S, McLean S, Hamilton B, Thomson A, Nair M and Rollins C: “The US/UK World Magnetic Model for 2010–2015,” Boulder, Colorado, USA: US NOAA technical report, National Environmental Satellite, Data, and Information Service/National Geophysical Data Center, 2010.

22. For more on the International Geomagnetic Reference Field (IGRF) model: Glassmeier K-H, Soffel H and Negendank JFW (eds): Geomagnetic Field Variations. Berlin: Springer-Verlag, 2009, http://www.ngdc.noaa.gov/IAGA/vmod/igrf.html (accessed July 21, 2013).

23. For more on the BGS Global Geomagnetic Model (BGGM): “BGS Global Geomagnetic Model,” British Geological Survey, http://www.geomag.bgs.ac.uk/data_service/directionaldrilling/bggm.html (accessed July 16, 2013).

Macmillan S, McKay A and Grindrod S: “Confidence Limits Associated with Values of the Earth’s Magnetic Field Used for Directional Drilling,” paper SPE/IADC 119851, presented at the SPE/IADC Drilling Conference and Exhibition, Amsterdam, March 17–19, 2009.

24. For more on the Enhanced Magnetic Model (EMM): Maus S: “An Ellipsoidal Harmonic Representation of Earth’s Lithospheric Magnetic Field to Degree and Order 720,” Geochemistry Geophysics Geosystems 11, no. 6 (June 2010): Q06015-1–Q06015-12.

For more on the High-Definition Geomagnetic Model (HDGM): Maus S, Nair MC, Poedjono B, Okewunmi S, Fairhead D, Barckhausen U, Milligan PR and Matzka J: “High Definition Geomagnetic Models: A New Perspective for Improved Wellbore Positioning,” paper IADC/SPE 151436, presented at the IADC/SPE Drilling Conference and Exhibition, San Diego, California, March 6–8, 2012.

25. Brooks AG, Gurden PA and Noy KA: “Practical Application of a Multiple-Survey Magnetic Correction Algorithm,” paper SPE 49060, presented at the SPE Annual Technical Conference and Exhibition, New Orleans, September 27–30, 1998.

Lowdon RM and Chia CR: “Multistation Analysis and Geomagnetic Referencing Significantly Improve Magnetic Survey Results,” paper SPE/IADC 79820, presented at the SPE/IADC Drilling Conference, Amsterdam, February 19–21, 2003.

Chia CR and de Lima DC: “MWD Survey Accuracy Improvements Using Multistation Analysis,” paper IADC/SPE 87977, presented at the IADC/SPE Asia Pacific Drilling Technology Conference and Exhibition, Kuala Lumpur, September 13–15, 2004.

26. Nyrnes E, Torkildsen T and Wilson H: “Minimum Requirements for Multi-Station Analysis of MWD Magnetic Directional Surveys,” paper SPE/IADC 125677, presented at the SPE/IADC Middle East Drilling Technology Conference and Exhibition, Manama, Bahrain, October 26–28, 2009.

27. For a detailed description of crustal magnetic modeling, including construction of the vector crustal magnetic field using downward continuation and trilinear interpolation: Poedjono B, Adly E, Terpening M and Li X: “Geomagnetic Referencing Service—A Viable Alternative for Accurate Wellbore Surveying,” paper IADC/SPE 127753, presented at the IADC/SPE Drilling Conference and Exhibition, New Orleans, February 2–4, 2010.

40 Oilfield Review

for local crustal magnetic mapping, including land, marine or aeromagnetic surveys. Fortu-nately, the crustal magnetic field needs to be characterized only once in the life of the reser-voir. The disturbance field, however, varies rap-idly over time. Because data are available from magnetic observatories, surveying engineers are

able to incorporate disturbances caused by diur-nal solar activity and magnetic storms into survey data processing.

The technique of infield referencing (IFR) makes use of data from local magnetic surveys near a wellsite to characterize the crustal mag-netic field. Service companies have developed

extensions of this technique, incorporating remote observatory data to account for time vari-ations. Surveying engineers use these techniques to extend the main magnetic field model and pro-vide the best estimate of the local magnetic field, which is critical for geomagnetic referencing and multistation drillstring compensation. These techniques allow magnetic surveying even at high latitudes, where the local magnetic field exhibits extreme variations.

Schlumberger has introduced the geomagnetic referencing service (GRS) as a cost-effective alter-native to conducting gyroscopic surveys in real-time drilling applications.28 GRS provides accurate data on wellbore position and enables timely cor-rections to wellbore trajectory. Surveying engi-neers use a proprietary algorithm, a 3D crustal model and a time- and depth-varying geomagnetic reference to correct MWD measurements for mag-netic drillstring interference, calculate tool orien-tation from the corrected measurements and advise the directional driller on course adjust-ments. Coordination between the operator, direc-tional drilling contractor, MWD survey provider, geomagnetic observatory and survey engineer is essential for managing this survey technique. Examples from the USA, Canada, offshore Brazil and offshore Ghana illustrate a range of geomag-netic referencing applications.

Avoiding Collision in the Marcellus ShalePennsylvania General Energy (PGE) has under-taken field development in the Marcellus Shale that illustrates the benefits of multiwell planning and the need for quantifying positional uncer-tainty and assuring collision avoidance. PGE and its service providers sought to optimize pad design for multiwell drilling.29 Historically, opera-tors have developed the Marcellus Shale and other resources in the Appalachian basin using inexpensive vertical wells with minimal quality control on well surveys conducted by gyro and steering tools. Currently, however, more opera-tors are turning to multiwell pads and horizontal drilling to improve logistics and economic and environmental impact during the development of shale gas reservoirs.

Operators now are drilling up to 14 wells per pad on 7-ft [2-m] centers by constructing devi-ated wells. First, a 171/2-in. surface hole is air drilled to a depth of about 1,000 ft [300 m] and then surveyed. A 12 1/4-in. section for a water pro-tection string is then air drilled to a true verti-cal depth (TVD) of 2,500 ft [760 m] using gyro-while-drilling tools to guide the separation of wells on the pad. The directional driller uses

> Plan view of wellbore trajectories, looking down. PGE used a multiwell pad design for 14 wells drilled into the Marcellus Shale from a single pad. The plan shows initial uncertainty disks at true vertical depths of 2,500 ft (red) and 5,000 ft (yellow). As expected, uncertainty grows larger with increasing distance from the surface location and can impact the drilling program. None of the red disks intersect each other, nor do the yellow disks, indicating that the wellbores (blue) are clear of each other at those depths. (Copyright 2010, SPE Eastern Regional Meeting. Reproduced with permission of SPE. Further reproduction prohibited without permission.)

> Pad design and well trajectories. PGE drilled 14 wells into two reservoirs during Phases 1 (magenta) and 2 (blue) of the drilling campaign. The graphical size of each wellbore corresponds to the size of the EOUs as defined in the survey program. The drilling team confirmed the anticollision condition. At the reservoir entry point, each well needed to have a minimum 200-ft [60-m] separation from its counterpart drilled in the opposite direction. (Copyright 2010, SPE Eastern Regional Meeting. Reproduced with permission of SPE. Further reproduction prohibited without permission.)

Autumn 2013 41

a north-seeking gyro until the well reaches a depth that is free of external magnetic interfer-ence from nearby wellbores. The deeper, devi-ated 8 3/4-in. section is simultaneously drilled and surveyed to total depth (TD) with a rotary steerable system (RSS) and MWD.

Because accurate surveying and anticollision monitoring are imperative, PGE took a proactive approach to the multiwell pad design and drilling by using a recently proposed anticollision stan-dard.30 Following this procedure, the operator defined uncertainty areas at three TVDs: 1,000 ft, 2,500 ft and 5,000 ft [1,500 m]. Well planners per-formed anticollision analysis of trajectories to ensure wellbores were properly separated at these depths. Visualization of wellbore trajecto-ries, with uncertainty areas plotted at intermedi-ate and deeper depths, confirmed that the drilling plan was unlikely to lead to wellbore col-lision (previous page, top).

The selection of slots in the multiwell pad was an important aspect of the PGE pad design because of the constraints on surface hole loca-tions and target coordinates. PGE drilled seven wells into each of two stacked reservoirs. The drilling engineer completed the final pad design after surface holes were drilled and surveyed; then they replanned all wells, recalculated uncertainty areas and reassessed anticollision conditions (previous page, bottom). As a result, the plan reduced the risk of wellbore collision and its associated costs.

Reaching Difficult Targets Offshore CanadaGeomagnetic referencing techniques have helped an operator efficiently and safely reach its objec-tives in the Jeanne d’Arc basin offshore eastern Canada.31 Weather conditions are often severe in this remote area of the North Atlantic, leading operators to develop strategies for minimizing the extent of their offshore installations. The con-struction of multiple extended-reach wells drilled from slots on gravity-based platforms leverages the use of infrastructure but creates a crowded sub-surface, placing a premium on collision avoidance and precise wellbore positioning.

As a further challenge, the geology of the area is complex. The sedimentary basin consists of thick, layered sandstones separated by shales and subdivided by faults into large compartments or blocks. The reservoir is in a fault-bounded sector in which the target zones are smaller than the seis-mic resolution. The operator needed to employ sophisticated drilling and surveying techniques to hit these small targets while maintaining tight restrictions on wellbore trajectory designs.

For a successful drilling program, the opera-tor required an accurate description of positional uncertainty and a small error ellipsoid. The GRS guided drilling program met these requirements and provided extended drillability, reduced drill-ing time and improved chances of hitting the geo-logic target (above).

High Precision in High LatitudesGeomagnetic referencing brings significant advantages but encounters its greatest challenge when applied at high latitude, where the magni-tude of geomagnetic disturbance field variations

> Hitting distant targets with an extended-reach well in the Jeanne d’Arc basin, offshore Canada. This well trajectory (center) extends approximately 7,000 m [23,000 ft] before dropping to hit two targets (red) at about 4,000 m [13,000 ft]. Insets (top and bottom) show close-up views of the targets and the ellipsoids of uncertainty (EOUs) for two survey methods. The positional uncertainty (green) of the magnetic surveys without GRS (top) is so large that the well may be outside the targets. With GRS (bottom), the positional uncertainty (blue) is well within the size of the targets. (Adapted from Poedjono et al, reference 27. The images in this figure are copyright 2010, IADC/SPE Drilling Conference and Exhibition. Reproduced with permission of SPE. Further reproduction prohibited without permission.)

0 m

–400 m

–800 m

–1,200 m

–1,600 m

–2,000 m

–2,400 m

–2,800 m

–3,200 m

–3,600 m

6,000 m

6,000 m5,000 m

4,000 m3,000 m

2,000 m

1,000 m

0 m

5,000 m

4,000 m

3,000 m

0 m

–400 m

–800 m

–1,200 m

–1,600 m

–2,000 m

–2,400 m

–2,800 m

–3,200 m

–3,600 mX-axis

Y-axis

Y-axis

Z-axis

28. Lowdon and Chia, reference 25. 29. Poedjono B, Zabaldano J, Shevchenko I, Jamerson C,

Kuntz R and Ashbaugh J: “Case Studies in the Application of Pad Design Drilling in the Marcellus Shale,” paper SPE 139045, presented at the SPE Eastern Regional Meeting, Morgantown, West Virginia, USA, October 12–14, 2010.

Kuntz R, Ashbaugh J, Poedjono B, Zabaldano J, Shevchenko I and Jamerson C: “Pad Design Key for Marcellus Drilling,” The American Oil & Gas Reporter, 54, no. 4 (April 2011): 111–114.

30. Poedjono B, Lombardo GJ and Phillips W: “Anti-Collision Risk Management Standard for Well Placement,” paper SPE 121040, presented at the SPE Americas E&P Environmental and Safety Conference, San Antonio, Texas, USA, March 23–25, 2009.

31. Poedjono et al, reference 27. Kuntz et al, reference 29.

42 Oilfield Review

is large. The Eni US Operating Co. Inc. Nikaitchuq field in the Beaufort Sea off the North Slope of Alaska, USA, is one such location. Continuity of the reservoir is broken by several faults, and drill-ers need to consider local reservoir compartmen-talization in well planning.32 Wellbore positioning must be precise and accurate.

At these high latitudes, the external distur-bance field varies dramatically over time.33 This disturbance represents the major source of noise in magnetic data used for well guidance. Amplitude variations are as large as 1,000 nT, and measured declination angles may vary by several degrees during magnetic storms. To account for these perturbations, GRS applies time-varying reference data from a nearby observatory to MWD measurements.

In 2009, the USGS launched a joint public-private partnership with Schlumberger to begin planning for installation and maintenance of a new observatory, called Deadhorse Geomagnetic Observatory (DED), at the town of Deadhorse, on the North Slope of Alaska. The newest of the 14 observatories, DED is now operated by Schlumberger under USGS guidance and follows Intermagnet standards.34

Instrumentation at the observatory includes a triaxial fluxgate magnetometer for vector field measurements, an Overhauser magnetometer for total field intensity measurements and a single-axis fluxgate declination-inclination mag-netometer (DIM) on a nonmagnetic theodolite. Specialists use DIM and Overhauser data to cali-brate the fluxgate variational data weekly. USGS scientists have developed specialized data pro-cessing algorithms to produce adjusted and defin-itive versions of real-time data streams received remotely at the USGS Geomagnetism Program headquarters in Golden, Colorado, USA.35

The workflow for geomagnetic referencing includes simultaneous acquisition and quality control of two data streams—MWD survey data at the rig site and real-time magnetic data at the observatory (above left).36 Schlumberger wellsite engineers perform QC of the raw MWD data. USGS experts execute automated QC and daily inspection of data from the DED observatory and apply sensor calibration factors to produce adjusted observatory data representing the time-varying disturbance field correction. GRS pro-cessing combines the time-stamped disturbance field data, crustal field data and main magnetic field model data. The algorithm applies the com-bined magnetic field data to the raw MWD sensor data at each survey depth and performs multista-tion processing and geomagnetic referencing,

> Geomagnetic referencing workflow. The workflow starts with raw MWD and magnetic observatory data streams (shown here as from the DED observatory) and combines them with crustal magnetic field data then progresses through geomagnetic processing, data adjustment and quality control. Processing continuously generates directional drilling corrections and provides definitive surveys at the end of bit runs. (Adapted from Poedjono et al, reference 32.)

Start

Real-time raw MWDsurvey data

QA/QC by Schlumberger

QA/QC•remove data with

external interference

QA/QC•calibration?•failed sensor?

Geomagneticreferencingprocessing

DED observatory•adjusted data•QA/QC by USGS

Crustal data•cube coordinates

Real-time GRS•azimuth correction

to drill ahead

Final GRS•definitive surveys•final GRS report

Sectional GRS•definitive surveys

PassQA/QC?

Stop

No

No

No

No

Yes

Yes

Yes

Yes

New bit run

End ofbit run?

Total depthreached?

Nearby well?

> Time-varying reference data. Raw magnetic MWD survey data (top, blue) initially exceeded the data quality acceptance limits (red) at several depths, but the data passed when referenced to DED observatory data (bottom). Initial acceptance limits were based on a static reference value (top, green) for the local magnetic field strength, whereas the DED data provided actual time-varying values (bottom, green) to which the limits could be referenced. (Adapted from Poedjono et al, reference 32.)

57,03910,500 11,776 13,052 14,328 15,605

Depth, ft16,881 18,157 19,433 20,710

10,500 11,776 13,052 14,328 15,605

Depth, ft16,881 18,157 19,433 20,710

DataReference

DataReference

57,239

57,439

57,639

57,839

58,039

58,239

nT

56,744

57,091

57,438

57,785

58,132

58,479

58,826

nT

Autumn 2013 43

32. Poedjono B, Beck N, Buchanan A, Brink J, Longo J, Finn CA and Worthington EW: “Geomagnetic Referencing in the Arctic Environment,” paper SPE 149629, presented at the SPE Arctic and Extreme Environments Conference and Exhibition, Moscow, October 18–20, 2011.

33. Merrill et al, reference 13.34. For more on Intermagnet: “International Real-time

Magnetic Observatory Network,” INTERMAGNET, http://www.intermagnet.org/index-eng.php (accessed October 16, 2013).

yielding geographic hole orientation. During additional processing stages, the algorithm implements data acceptance logic and computes a correction to drilling direction. The directional driller applies the drill-ahead correction until a new set of surveys is completed and a new drill-ahead correction is available. At the completion of each BHA run, surveying engineers apply BHA deflection corrections and compile the final definitive survey for that run.

The use of time-varying reference data was essential for drilling engineers to plan and exe-cute drilling in the Nikaitchuq field. Magnetic MWD survey raw data initially failed the data quality acceptance limits at several depths but improved to an acceptable range when refer-enced to DED observatory data (previous page, bottom). Because the company used GRS, drill-ing activities continued without the need for dedicated and costly surveying operations beyond the standard MWD survey stations.

High-Density Wells in the Williston Basin ConocoPhillips Company has demonstrated that improved wellbore survey accuracy contributes to increased oil production. Better survey accu-racy enables closer well separation and longer horizontal wells essential for boosting the effi-ciency of water injection programs designed to enhance oil recovery. Operating in two fields near the Cedar Creek anticline along the border between Montana and North Dakota, USA, the company has systematically studied the accuracy of existing wellbore survey data and examined the causes of MWD errors. By developing improved methodologies for magnetic data col-lection and reducing those errors, the company reduced positional uncertainty and contributed to both the safety and viability of the horizontal drilling program.37

Initially, the operators in these fields placed horizontal wells on a 640-acre [2.6-km2] spac-ing. They subsequently reduced well spacing to 320 acres [1.3 km2] and reconfigured the well pattern for a line-drive waterflood, in which rows of injection wells alternated with rows of producers (above right). Reservoir modeling suggested that reducing well spacing to 160 acres [0.65 km2] would be beneficial. However, before proceeding, the operator needed to assess the accuracy of wellbore place-ment, because inadvertent convergence of bore-holes could adversely affect waterflood sweep efficiency, reducing hydrocarbon production and increasing lifting and disposal costs.

To assess the accuracy of MWD surveys, the operator conducted several statistical surveys in which the positions of wells drilled using MWD were compared with positions determined from postdrilling gyro surveys. Results showed that while the average azimuth deviation between the MWD and gyro data was about 1°, the differences were larger for a significant number of wells. After evaluating the data, surveying engineers

determined that the principal cause of azimuthal error was BHA-induced magnetic interference. Other factors included local magnetic field varia-tions and drillstring sag.

Understanding and minimizing BHA-induced magnetic interference proved to be the key to improving survey accuracy. Surveying engineers used specialized software to estimate the contri-bution of drillstring interference to azimuth error

> Field development plan. In a field in Montana and North Dakota, USA, operators started field development with one well per 1-mi2 [640-acre, 2.6-km2] parcel. Alternate rows of injector (blue) and producer wells (gray) show planned down-spacing to an interwell spacing of 950 ft [290 m] (red box). Positional uncertainty needs to be minimized to keep the well trajectories parallel and reduce the risk of premature breakthrough of water from the injectors. (Adapted from Landry et al, reference 37.)

950 f

t

1 mi

1 m

i

Early producer well spacing:one well per 640 acres

Williston Basin

35. Love and Finn, reference 19.36. For more on the workflow at the DED observatory and on

the geomagnetic referencing: Poedjono et al, reference 32.

37. Landry B, Poedjono B, Akinniranye G and Hollis M: “Survey Accuracy Extends Well Displacement at Minimum Cost,” paper SPE 105669, presented at the 15th SPE Middle East Oil and Gas Show and Conference, Manama, Bahrain, March 11–14, 2007.

44 Oilfield Review

and evaluate the benefits and tradeoffs of placing nonmagnetic material between the magnetome-ters and the rest of the BHA. Because separating sensors from the bit can compromise real-time steering, the operators minimized nonmagnetic components and instead employed single station and multistation processing techniques to correct the surveys in real time. Postdrilling comparisons of MWD drilled trajectories with gyro surveys con-firmed that discrepancies had been reduced sta-tistically, including for instances in which the real-time magnetic interference corrections were large. Taking the reduced EOUs into account, drilling engineers were able to stagger wellhead positions and optimize wellhead spacing to pre-vent water breakthrough (left).

Crustal Variations In some situations, the main concern is not the time-varying field but the crustal correction. Such was the case for one operator in a deepwa-ter heavy-oil field offshore Brazil.38 The project lies in 1,100 m [3,600 ft] of water in the northern Campos basin. The operator had drilled several wells using MWD and had observed discrepancies between downhole tool readings and those expected from the BGGM. To improve magnetic surveying here, it was necessary to develop a bet-ter model of the local magnetic field so that well-bore trajectories would attain their targets. The company needed to employ a highly accurate geo-magnetic model to avoid field acceptance criteria failures in real-time drilling. Such failures may lead to unnecessary tool retrieval operations because of suspected tool failure.

To resolve the survey discrepancies, a research team composed of representatives from the operator, Schlumberger, other contractors and academia developed a method for mapping the magnetic variations using the High-Definition Geomagnetic Model (HDGM2011), which had recently been developed at the US NGDC. The team integrated this large-scale magnetic field model with data from a local aeromagnetic sur-vey to extend the spatial spectrum of the mag-netic field from regional scales down to the kilometer scale (left).

The team used two independent methods to analyze the crustal magnetic model.39 Method 1 combined the BGGM with aeromagnetic survey data and employed an equivalent source method for downward continuation of the field to reser-voir depth. Method 2 combined the aeromagnetic

> Strategies for ensuring optimal spacing to prevent water breakthrough. Survey Program B (pink) delivers higher accuracy than Survey Program A (blue). Had Wells 1 and 2 been drilled from adjacent surface locations using Survey Program A, the wells may have collided at TD. Survey Program B, with compensation for magnetic interference, ensures noncollision and allows the wells to be extended to planned total depth. By staggering one wellhead to the surface location of Well 3, the operator could increase well separation at total depth, drill wells with the desired orientation and spacing and prevent early water breakthrough. The operator chose to use both Survey Program B and wellhead staggering. (Adapted from Landry et al, reference 37.)

Surface location Well 1

Surface location Well 3

Well-to-well separationat surface location

Survey Program Adoes not provide

separation at TD.

Separation atmeasured depth

Separation for SurveyProgram A relative to the offset at TD of Well 3

Survey Program B provides separation at TD.

Plan View

Surface location Well 2

Uncertainty of Survey Program A

Uncertainty of Survey Program B

>Magnetic field declination maps offshore Brazil. The standard model (left) shows smooth, large-scale variations in magnetic field declination in the vicinity of the hydrocarbon field (red polygon). The higher resolution HDGM (center) includes more detail. The combined HDGM and aeromagnetic survey model (right) contains the highest resolution information of all three models. All maps show declination at mean sea level. Differences of nearly 1° in declination are observed between the standard and highest resolution models near the field. (Adapted from Poedjono et al, reference 38.)

–23.2

–23.2

–23

–23.2

–22.2

–23.2

–23.6

–23

–22°48’

–22°24’

–22°00’

–21°36’

–21°12’

–20°48’

–40°00’ –39°36’–40°00’ –39°36’–40°00’ –39°36’

–23.4

–23.4 –23.4–23.8

Longitude

Field

Latit

ude

–23

–22

–24

Mag

netic

fiel

d de

clin

atio

n, d

egre

e

Autumn 2013 45

survey with a long-wavelength crustal field model provided by the German CHAMP satellite survey and created a 3D magnetic model for the lease area. The team established the validity of Method 2 by comparing the results with marine magnetic profiles taken from the US NOAA/NGDC archive. Magnetic field model attributes com-puted with these two methods closely agreed with each other when compared at mean sea level and at the 5,000-m [16,400-ft] reservoir depth (right).

The team discovered that intermediate-wave-length anomalies caused by large-scale magneti-zation of the oceanic crust had a significant impact on local magnetic declination. The higher-resolution geomagnetic reference models enabled more-refined multistation compensation for drillstring interference. By comparing predic-tions of horizontal and vertical magnetic field components with those from MWD tool readings, the team established validity of the broadband models. Data points affected by drillstring inter-ference were outside quality control acceptance bands when processed with the BGGM but were consistent with the other data when processed with a high-resolution model.

The team evaluated the importance of the time-varying disturbance field using data from the nearby Vassouras Magnetic Observatory in Brazil. Results showed small variations in decli-nation, dip and total field intensity. Diurnal varia-tions were insignificant at the wellbore positions during times of low solar activity, and data from the high-resolution static models were sufficient for these times. Operator representatives con-cluded that multistation analysis improved when they used the high-resolution geomagnetic mod-els compared with the BGGM magnetic field pre-dictions. Significant localization improvements occurred when they used GRS to correct MWD raw readings. Estimated wellbore bottomhole locations shifted significantly, and the sizes of the ellipsoids of uncertainty and the TVD uncer-tainty consistently decreased.

> Crustal contribution to the magnetic field declination at two depths in the vicinity of a field offshore Brazil. The crustal field contribution to the magnetic declination is shown in plan view at mean sea level (top) and at a depth of 5,000 m (bottom). Values were calculated using a method that combined an aeromagnetic survey with a long-wavelength crustal field model provided by the German CHAMP satellite survey; the method then created a 3D magnetic model for the lease area. The 3D magnetic field changes with depth, in large part because of the magnetic properties of the Earth’s crust underlying the sediments offshore Brazil. (Adapted from Poedjono et al, reference 38.)

2.0

1.8

1.6

1.4

1.2

1.0

0.8

0.6

0.4

0.2

0

–0.2

–0.4

–0.6

0

0

0.2

0.2

–0.2

0.4

0.4

0.6

0.6

0.6

0.8

0.8

1.2

1.4

Mag

netic

fiel

d de

clin

atio

n, d

egre

eM

agne

tic fi

eld

decl

inat

ion,

deg

ree

2.0

1.8

1.6

1.4

1.2

1.0

0.8

0.6

0.4

0.2

0

–0.2

–0.4

–0.6

10,000 m

10,0

00 m

10,000 m

10,0

00 m

Crustal Contribution at Sea Level

Crustal Contribution at 5,000 m

38. Poedjono B, Montenegro D, Clark P, Okewunmi S, Maus S and Li X: “Successful Application of Geomagnetic Referencing for Accurate Wellbore Positioning in a Deepwater Project Offshore Brazil,” paper IADC/SPE 150107, presented at the IADC/SPE Drilling Conference and Exhibition, San Diego, California, March 6–8, 2012.

39. Two proprietary processing methods developed for analyzing the crustal field are discussed in Poedjono et al, reference 38. Method 1 was developed by Fugro Gravity & Magnetic Services Inc, now part of CGG. Method 2 was developed by Magnetic Variation Services LLC.

46 Oilfield Review

Deepwater Success Accurate real-time magnetic surveys allow direc-tional drillers to stay on path and to reduce the number of required confirmatory gyro surveys. Tullow Ghana Ltd. used geomagnetic referencing to achieve its objectives to hit distant geologic targets accurately and within budget while devel-oping the Jubilee field offshore Ghana.40

The operator wanted to drill all wells safely and successfully in the shortest possible time because rig spread costs are exceptionally high in this area. To enable accurate GRS, Schlumberger surveying experts conducted numerical simulations, which quantified the sensitivity of

the magnetic measurement to wellbore trajectory and to the inclusion of nonmagnetic collars for BHA variations (above).

An aeromagnetic survey provided the basis for the custom-built geomagnetic model. This 80-km × 80-km [50-mi × 50-mi] survey was centered at the Jubilee field at an altitude of 80 m [260 ft] and included presurvey test flights for calibration and use of a base station as reference for time-varying changes in the magnetic field. Analysts computed a total magnetic intensity (TMI) anomaly grid using the total magnetic field measured in the aeromagnetic survey combined with the 2010 BGGM main magnetic field model.41 Crustal

magnetic field processing yielded an updated magnetic field from sea level to a depth of 4,500 m [14,800 ft] using downward continuation of the scalar TMI anomaly. Subsequent processing determined the east, north and vertical components of the magnetic field and transformed them into declination and inclination perturbations relative to the main magnetic field.

> Quantifying magnetic measurement sensitivity to toolstring interference. Modeling codes are used to simulate the extent of magnetic interference for various survey orientations and BHA designs. This simulation, taken from the Schlumberger Drilling & Measurements Survey Tool Box, shows the large azimuthal error (red) that would occur at this particular wellbore grid azimuth of 270° and inclination of 90° if the driller did not add nonmagnetic spacing material to the BHA in addition to that included in the initial design (blue). Drilling engineers use these simulations to determine the length of nonmagnetic material above or below the MWD measure point necessary to reduce the error sufficiently.

BHA Collar Size

Azimuth Reference

Nonmagnetic Spacing of BHA Elements

Units Survey Data Geomagnetic Reference

Schlumberger Drilling & Measurements Survey Tool Box

BHA Configuration Type

1

2

3

4

Results

Small (4.75” OD or Less)

MetersGridAzimuth

Inclination

Declination GeoMag Field Geographic Region

Grid Convergence Total lBl

RegionDEM Total CorrectionCheck Reference

Minimum Number of Surveys Required for DMAG

deg

deg

deg

deg

ft

ft ft

ft

Steel

NonMag

ft

ft ft

ft ft

degDip nT

Feet

True North

Grid North

Medium (6.75” and all Medium sizes)

PowerDrive Rotary Steerable

Steerable Motor Assembly

NonMag Steerable Motor

Drill Collars or Other BHA

Stabilizer + Bit Only Below MWD

Azimuth Error:

Add Nonmagnetic Spacing

4.22

270.00 4.5900

0.2100

4.3800

49895 61.7590.00

25

D2

D2

L2 L1S2 S1

D1

D1

MP

MP

100.51 52.78 36.02

deg

NMR

deg

nT

nT

1746

330.0

0.00.07

Calculate

Open Save Save As .. Clear Exit

Report

Interfering Field:

Above MWD Add ..

Undo LastBelow MWD

Delta FAC dlBl

Delta FAC dDip

Large (8” OD or More)

Main

EDI Calculator Reference Check Benchmark Rotation Shot BHA Survey Frequency

File Launch Help

40. Poedjono B, Olalere IB, Shevchenko I, Lawson F, Crozier S and Li X: “Improved Drilling Economics and Enhanced Target Acquisition Through the Application of Effective Geomagnetic Referencing,” paper SPE 140436, presented at the SPE EUROPEC/EAGE Annual Conference and Exhibition, Vienna, Austria, May 23–26, 2011.

41. For more on the processing workflow: Poedjono et al, reference 32.

Autumn 2013 47

For the initial wells in the Jubilee field, stan-dard MWD surveys yielded small enough EOUs to hit the geologic targets with confidence. These initial well paths had relatively shallow inclination angles. For more-distant targets with higher inclination angles and longer tan-gent sections, the uncertainty associated with standard MWD surveys was unacceptably large. However, uncertainty was considerably smaller for GRS processed magnetic data, and drillers reached their objectives with high confidence. Using GRS, the operator was able to drill the well with guaranteed placement of the wellbore inside the target (above).

Reaching the TargetThese examples illustrate a range of new and exacting requirements for wellbore guidance and the geomagnetic measurement technology that has been developed to satisfy those require-ments. Challenges have included avoiding well-bore collision, reducing drillstring magnetic interference and accounting for geomagnetic field variations associated with crustal magne-tism and temporal magnetic field variations.

Directional drillers now place wellbores within increasingly demanding targets by relying on real-time wellbore surveys and small EOUs. High-resolution geomagnetic reference models

aid processing for drillstring interference com-pensation and enhance measurement quality control by employing customized acceptance criteria. Geomagnetic referencing improves well placement accuracy, reduces positional uncertainty and mitigates the danger of colli-sion with existing wellbores. When used in real-time wellbore navigation, GRS saves rig time, reduces drilling costs and helps drillers reach their targets. —HDL

> An extended-reach well in the Jubilee field offshore Ghana. The Tullow Ghana Ltd. Well 4 has a long step-out and tangent profile to hit the target (red). The EOU from standard MWD (top left , green) is larger than the rectangular geologic target. Because of the smaller EOU from GRS (center left , blue), the operator was able to drill the well with high confidence that the wellbore would penetrate the target. (Adapted from Poedjono et al, reference 40. The images in this figure are copyright 2011, SPE EUROPEC/EAGE Annual Conference and Exhibition. Reproduced with permission of SPE. Further reproduction prohibited without permission.)

0 m

0 –400 m

–400 –800 m

–800 –1,200 m

–1,200 –1,600 m

–1,600 –2,000 m

–2,000 –2,400 m

–2,400–2,800 m

–2,800 m–3,200 m

–3,200 m 515,000 m

515,000 m

514,000 m

514,000 m

513,000 m

513,000

512,000 m

512,000

X-axis

X-axisY-axis

Z-axis

X-axis

511,600 m

511,800 m

512,000 m

511,600 m

511,800 m

512,000 m

X-axis

Y-axis

48 Oilfield Review

Blowing in the Solar Wind: Sun Spots, Solar Cycles and Life on Earth

The Sun regularly experiences eruptions that shower space with energetic ions. In

1859, a massive solar event occurred with a magnitude that surpassed that of all other

recorded events, and the Earth was directly in the path of the storm. Hours after the

eruption, sparks began to fly from telegraph wires, fires were ignited by downed

wires, equipment operators felt electrical shocks from their telegraph keys and ticker

tapes burst into flames. A century and a half later, should a similar solar event occur,

more than wires and paper would be at risk.

Anatoly ArsentievIrkutsk, Russia

David H. HathawayNational Aeronautics and Space Administration (NASA) Marshall Space Flight CenterHuntsville, Alabama, USA

Rodney W. LessardHouston, Texas, USA

Oilfield Review Autumn 2013: 25, no. 3. Copyright © 2013 Schlumberger.For help in preparation of this article, thanks to Don Williamson.

Those of us in the energy industry owe our liveli-hoods to the Sun. The hydrocarbons we search for and produce were formed from organic matter that stored ancient energy that originated within the Sun. In the not too distant past, the Sun was an object of reverence because of its control over our lives. Today, familiarity with and understand-ing of the Sun has removed much of our sense of veneration; however, we understand that our very existence is based on a relationship to the seem-ingly unchanging presence of the solar system’s shining star.

On occasion, however, the Sun’s apparent sta-bility is interrupted by powerful displays of its dynamism. One such example occurred on the morning of September 1, 1859. From his private observatory, amateur astronomer Richard Carrington observed a cluster of large spots on the surface of the Sun. Suddenly, a brilliant flash of white light—a solar flare—erupted from the area of the spots.1 This particular flare was the harbinger of a gigantic coronal mass ejection (CME), which spewed solar plasma into inter-planetary space.

This massive cloud of charged particles arrived at Earth in less than 18 hours. It pro-ceeded to disrupt the most advanced technology of the day—the telegraph.2 The interaction between the CME and the Earth’s magnetic field induced electrical currents in exposed telegraph wires. Current raced through the wires, causing some of them to overheat, fall to the ground and set off fires. Telegraph machines were hit by pow-

erful surges of electricity, which administered electrical shocks to the operators. Some reports described telegraph paper bursting into flames and machines that continued to receive informa-tion, even after the operators had disconnected their battery power. Disturbances in the Earth’s magnetic field from the effects of the CME caused compass needles to behave erratically. The effects were seen not just on the Earth’s surface; auroras, which are normally restricted to Earth’s higher latitudes, lit the sky as far south as the Caribbean region.

Most experts consider the superstorm of 1859, referred to as the Carrington event, to be the largest recorded solar storm to directly impact the Earth. Data from ice cores dating back 500 years show evidence of geomagnetic storms of varying intensity, but none reached the magnitude of that singular episode.3

Modern infrastructure has become dependent on a multitude of interconnected systems and devices that are sensitive to electromagnetic and geomagnetic forces. Scientists are concerned that another Carrington-type CME directed toward Earth would wreak havoc, overwhelming electri-cal power grids and control systems, destroying telecommunications satellites, disrupting global positioning systems (GPSs) and plunging whole continents into darkness and disarray. In 1989, a much smaller geomagnetic storm caused a black-out that pitched the province of Quebec, Canada, into darkness and disrupted power in many loca-tions in the Northeast US.

1. Cliver EW: “The 1859 Space Weather Event: Then and Now,” Advances in Space Research 38, no. 2 (2006): 119−129.

2. Boteler DH: “The Super Storms of August/September 1859 and Their Effects on the Telegraph System,” Advances in Space Research 38, no. 2 (2006): 159−172.

3. Stephens DL, Townsend LW and Hoff JL: “Interplanetary Crew Dose Estimates for Worst Case Solar Particle Events Based on Historical Data for the Carrington Flare of 1859,” Acta Astronautica 56, no. 9−12 (May−June 2005): 969−974.

Autumn 2013 49

50 Oilfield Review

According to solar scientists, predicting the next Carrington-class event, or any solar storm, is practically impossible. When solar flares and CMEs occur, scientists have found it difficult to determine whether the Earth lies directly in the path of these streaming ions. In the past few years, the ability to issue alerts about potential damaging solar storms has been improved by the deployment of satellites strategically positioned to monitor the Sun’s activity.

Although scientists are not able to forecast exactly when solar flares and CMEs will occur, they have discovered a correlation between an increase in the number of sunspots and the fre-quency and intensity of solar events. Sunspots are dark regions on the Sun, and they follow an 11-year cycle. During sunspot cycle minima, there may be no visible spots; during maxima the number may be greater than 200. Each cycle is numbered, dating to 1755, when observers began to systematically record sunspot activity (above). The Carrington event occurred at the peak of Cycle 10. The US National Oceanic and Atmospheric Administration (NOAA) Space Weather Prediction Center (SWPC) predicts that Cycle 24 will peak in 2013.4 The Sun has been relatively quiet during Cycle 24, but the potential always exists for the Sun to unleash another Carrington-like event.

This article discusses the concepts of solar cycles, solar events, CMEs, space weather, solar monitoring and the potential effects of solar storms on modern infrastructure, and it reviews current warning systems.

A Not So Benign SunAbout 5 billion years ago, a cloud of dust and gas approximately 1.6 trillion km [1 trillion mi] in diameter coalesced to form our solar system.5 The source of that cloud is believed to be a mix of pri-mordial gas and material from older stars that exploded in massive supernovae.6 Gravity collapsed the cloud in upon itself, and mutual attraction of the particles accelerated the collapse to form a dense central core. Rotation of the cloud acceler-ated with contraction, while centrifugal forces flat-tened the cloud toward the edges, leaving a bulge near the center from which the Sun evolved.

As the central core of the Sun continued to col-lapse, the compression generated heat, which melted and vaporized the dust. About 10 million years after the collapse began, the rate of collapse slowed because the pull of gravity was balanced by the pressure of hot gases. The rising core tempera-ture initiated nuclear reactions, and the heat and pressure stripped away electrons, leaving mostly plasma—a mixture of protons and electrons. The gravitational pull of the Sun continued to com-press the plasma in its core to densities nearly ten times that of lead and heated the plasma to nearly 16 million °C [29 million °F], at which point fusion reactions can occur.

In the Sun’s fusion reaction, hydrogen atoms fuse and form helium. During the reaction, some of the original mass is converted into heat and photons. The photons radiate outward, first trav-eling through the radiative zone and then, after millions of collisions, arrive at the region near

the surface—the convection zone (next page, top right). From the convection zone, the photons eventually leave the Sun. Traveling at the speed of light, photons cover the 150 million km [93 mil-lion mi] between the Earth and Sun in about eight minutes.

The photons emitted by the Sun cover a broad band of the electromagnetic spectrum—from high-energy X-rays to radio waves. The Earth is constantly bombarded by this energy, but because the atmosphere shields it from most of the emis-sions, only a few specific frequencies—mostly those of ultraviolet light, visible light and radio waves—reach Earth’s surface.

A self-generated magnetic field is a by-product of the Sun’s fusion reactor, rotation and con-stantly moving mass of plasma in the convection zone. Magnetic field lines are generally aligned with the axis of rotation of the Sun. The field exhibits a dipolar nature analogous to that of the Earth, with its north and south magnetic poles. However, unlike Earth’s magnetic field, the Sun’s magnetic field reverses polarity on a regular basis, coinciding with the midpoint of the 11-year sunspot cycle peak.

The Sun’s rotating magnetic field also gener-ates a current sheet that extends billions of kilo-meters from the Sun out into space. When the magnetic polarity reversal occurs—a process that started in the summer of 2013 for Cycle 24—the current sheet becomes highly contorted. The Earth dips in and out of the current sheet while orbiting the Sun, potentially creating stormy space weather conditions.7

> Sunspot cycles. Scientists have systematically recorded the number of sunspots and numbered the sunspot peaks dating from the 1700s. In several recent cycles, sunspot counts approached or exceeded 200; the current cycle average count is less than 100.

0

50

100

150

200

250Su

nspo

t num

ber

Date1750

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

1770 1790 1810 1830 1850 1870 1890 1910 1930 1950 1970 1990 2010

Carrington event

Quebec blackout

Autumn 2013 51

On the surface of the Sun, magnetic field lines emerge to form sunspots. Magnetic field lines may encompass volumes that are quite large—the planet Jupiter, which is 150,000 km [90,000 mi] in diameter, could easily fit inside some of them (below right). Coronal loops also form at the surface, following the magnetic field lines. During solar sunspot peaks, the number of coronal loops increases and magnetic field lines often become twisted. This twisting stores mas-sive amounts of energy that is eventually released in the form of solar flares, CMEs and other events. Space weather is punctuated by bursts of energy from these magnetic disturbances.

Space WeatherSpace weather is defined as the physical condi-tions in the space environment that have the potential to affect space-borne or ground-based technology systems.8 Space weather is greatly influenced by the energy carried from the Sun by the solar wind, and it can disturb conditions immediately around the Earth. Charged parti-cles—mainly protons and electrons—make up the solar wind. These particles are emitted in all directions from the Sun. Solar wind speed, den-sity and composition determine associated effects on the Earth.9 Geomagnetic storms, iono-spheric disturbances and aurora emissions are all manifestations of space weather. Coronal mass ejections and associated shock waves are the most violent components of space weather, and they tend to compress the Earth’s magneto-sphere and trigger geomagnetic storms.

Earth’s magnetosphere is a bullet-shaped bubble that protects the planet’s surface from harmful radiation. The magnetosphere shields the Earth from fast-moving ions by deflecting and concentrating them at the Earth’s north and south poles. The Van Allen radiation belts trap charged particles that leak through the magnetosphere, further protecting Earth’s sur-face from harmful electromagnetic radiation.

4. “NOAA: Mild Solar Storm Season Predicted,” National Oceanic and Atmospheric Administration (May 8, 2009), http://www.noaanews.noaa.gov/stories2009/20090508_solarstorm.html (accessed September 4, 2013).

5. Friedman H: The Astronomer’s Universe: Stars, Galaxies and Cosmos. New York City: Ballantine Books, 1991.

6. Naturally occurring heavy elements found on Earth, such as uranium and plutonium, could have come only from an extremely violent nuclear reaction such as a supernova.

7. Phillips T: “The Sun’s Magnetic Field Is About to Flip,” NASA (August 5, 2013), http://www.nasa.gov/content/goddard/the-suns-magnetic-field-is-about-to-flip (accessed August 28, 2013).

8. Hanslmeier A: The Sun and Space Weather, 2nd ed. Dordrecht, The Netherlands: Springer, 2007.

9. Feldman U, Landi E and Schwadron NA: “On the Sources of Fast and Slow Solar Wind,” Journal of Geophysical Research 110, no. A7 (July 2005): A07109.1–A07109.12.

> The Sun’s structure. Fusion reactions take place in the Sun’s central core. The pull of gravity accelerates hydrogen nuclei inward, toward the Sun’s center, where they fuse and form helium; the reaction releases energy. The energy—in the form of photons and other elementary particle by-products—rises through the Sun’s radiative and convection zones and then exits from the photosphere. The corona is the Sun’s outer atmosphere, a layer of plasma surrounding the chromosphere. Features displayed on the Sun’s surface seen here include a prominence, solar flares, sunspots and a coronal hole. [Illustration courtesy of the US National Aeronautics and Space Administration (NASA).]

Flare

Chromosphere

Sunspots

Photosphere

Convection zoneRadiative zoneCore

Prominence

Coronal hole

Corona

> The Sun’s magnetic field lines. Convoluted magnetic field lines (green) may extend thousands of kilometers out from the surface of the Sun. (Image courtesy of the NASA Goddard Space Flight Center Scientific Visualization Studio.)

52 Oilfield Review

The region of the magnetosphere away from the Sun is elongated by the pressure of the solar wind, and the shape varies with space weather conditions (left).

Space weather has the potential to cata-strophically disrupt the near-Earth environment. The World Meteorological Organization (WMO), an agency of the United Nations, established the Interprogramme Coordination Team on Space Weather (ICTSW) to address concerns of poten-tial disruptions to life on Earth caused by space weather.10 Experts from twenty countries and seven international organizations participate in the program. In the US, NOAA is responsible for monitoring terrestrial as well as space weather. The NOAA SWPC constantly monitors data about the Sun and forecasts solar and geophysical events that may impact satellites, navigation sys-tems, power grids, communications networks and other technology systems.11 Because of the correlation of increases in sunspot numbers with solar storms, scientists are on high alert during solar maxima.

SunspotsAbout 2,800 years ago, Chinese astronomers made the first recorded observation of sunspots.12 The invention of the telescope in the 1600s made it possible to study and record the ever-changing face of the Sun more closely. Reliable and system-atic records of sunspots date back to the 1700s. In the mid-1800s, German astronomer Samuel Heinrich Schwabe first identified a 10-year pat-tern of the rise and fall of sunspots—the sunspot cycle. Swiss astronomer Johann Rudolf Wolf later characterized the 11-year period for the cycle and developed a formula for quantifying sunspot activity, the Wolf number, which is still in use today.13 The cycle is not exactly 11 years but has varied from 9 to 14 years.

Sunspots form where concentrated magnetic field lines project through the hot gases of the photosphere and correspond to regions that are cooler than the surrounding surface. Although they appear darker than the rest of the solar disk, removed from the Sun, they would be brighter than anything else in the solar system (left). The importance of the complex magnetic fields to the activity of the Sun has been realized only within the past 100 years. American astronomer George Ellery Hale first reported solar magnetism in 1908. He determined the presence of magnetic fields by measuring changes to intensity and polarization of light emitted from atoms in the Sun’s atmosphere.14 Hale and his colleagues dem-onstrated that sunspots contain strong magnetic

> Earth’s magnetosphere. The magnetosphere, the area of space around the Earth created by Earth’s magnetic field, is a dynamic structure that responds to variations in solar activity and space weather. Solar wind, which compresses the sunward side of the magnetosphere, determines its shape. A supersonic shock wave—the bow shock—forms on the sunward side of Earth. Most of the solar wind particles are slowed at the bow shock and directed around the Earth in the magnetosheath. The solar wind pulls at the magnetosphere on the Earth’s night side, extending the length of the magnetosphere up to 1,000 Earth radii, creating what is known as the magnetotail. The outer boundary of Earth’s confined geomagnetic field is called the magnetopause. Trapped charged particles—the Van Allen radiation belts, the plasmasphere and the plasma sheet—reside within the magnetosphere. (Adapted from an image courtesy of Aaron Kaase, NASA Goddard Space Flight Center.)

Interplanetary magnetic field lines

Bow shock

Solar wind

Magnetosheath

Magnetopause

Plasma sheet

Plasmasphere

Van Allenradiation belts

Magnetotail

> Sunspots. Regions on the Sun that appear darker than the rest of the disk, sunspots are formed by concentrated magnetic fields that project through the hot gases of the photosphere out to the Sun’s surface. These magnetic fields create cooler, darker regions called sunspots. The dark center of a sunspot is called the umbra; the light area around the umbra is the penumbra. Sunspots occur in groups and frequently in pairs. The two spots in a pair have opposite magnetic polarities. (Photographs courtesy of NASA.)

Umbra

Penumbra

Penumbra

Umbra

Autumn 2013 53

10. For more on WMO and ICTSW: “WMO Scientific and Technical Programs,” World Meteorological Organization, http://www.wmo.int/pages/prog/ (accessed August 1, 2013).

11. For more on the SWPC: NOAA National Weather Service Space Weather Prediction Center, http://www.swpc.noaa.gov/AboutUs/index.html (accessed August 13, 2013).

12. Clark DH and Stephenson FR: “An Interpretation of the Pre-Telescopic Sunspot Records from the Orient,” Quarterly Journal of the Royal Astronomical Society 19, no. 4 (December 1978): 387−410.

13. Hathaway DH: “The Solar Cycle,” Living Reviews in Solar Physics 7 (2010): 1–65.

14. Alexander D: The Sun. Santa Barbara, California, USA: Greenwood Press, 2009.

fields and that all the sunspot groups in a given solar hemisphere have the same magnetic polar-ity signature. Furthermore, sunspot polarity cor-relates to the Sun’s magnetic field orientation in a specific solar cycle, which reverses with each cycle. The hemisphere that has a north magnetic polarity at one solar minimum has a south mag-netic polarity at the next solar minimum.

Sunspots typically range in size from 2,500 to 50,000 km [1,500 to 30,000 mi] and cover less than 4% of the Sun’s visible disk. In comparison, the Earth’s diameter is about 12,700 km [7,900 mi]. Sunspots typically have a lifetime of a few days to a few weeks and tend to be concen-trated in two midlatitude bands on either side of the Sun’s equator. During the early part of the solar cycle, sunspots are most commonly seen around latitudes of 25° to 30° north and south of the equator. Later in the cycle, they appear at latitudes of 5° to 10°. Sunspots rarely occur at latitudes above 50°.

The intense magnetic fields associated with sunspots often create arching columns of plasma called prominences that appear above sunspot regions (right). Some prominences may hang sus-pended above the solar surface for several days. When these massive loops of energy become twisted, they store energy that can violently erupt and blast coronal material outward from the Sun as a solar flare or a CME.

Solar Flares and CMEsThe energy source for solar flares originates in the tearing and reconnecting of magnetic field lines, and the strong magnetic fields in active sunspot regions often give rise to solar flares (right). These intense, short-lived releases of energy are our solar system’s most explosive events. During a solar flare, temperatures soar to 5 million °K, and vast quantities of particles and radiation can be blasted into space, but a flare usually ends within 20 minutes.

> Solar prominence photographed on September 23, 1999. The space-based Solar and Heliospheric Observatory (SOHO) captured this image of an eruptive prominence using extreme ultraviolet frequencies. The release of energy from twisted magnetic field lines flings plasma above the Sun’s surface. [Photograph courtesy of the SOHO Extreme Ultraviolet Imaging Telescope (EIT) consortium.]

Prominence

> Solar flare. The NASA Solar Dynamics Observatory (SDO) captured this image of a solar flare on May 22, 2013. The image captures light in the 13.1-nm wavelength, which highlights material heated to intense temperatures during a flare. The teal coloration is typical of images using this wavelength. (Photograph courtesy of the NASA SDO.)

54 Oilfield Review

During the peak of the sunspot cycle, several flares may occur daily. When a flare erupts, ultra-violet and X-ray radiation from the flare travel at the speed of light, arriving at the Earth in about

8 minutes. A day or two later, high-energy parti-cles may also arrive at the Earth, producing auro-ras—lights in the polar night skies—and affecting radio communications (above).15

During some solar flares, a more violent reac-tion may occur—a coronal mass ejection (below left). When the twisted magnetic field lines cross, their stored energy explodes outward with tre-mendous force. A CME occurs when the force of the released energy flings a mass of superheated plasma from the Sun’s surface into space.

CMEs vary in intensity and magnitude. A large CME can contain 9 × 1012 kg [20 × 1012 lbm] of matter that may be accelerated into space at sev-eral million kilometers per hour. The speed at which the plasma travels depends on the original energy release. A high-energy CME can arrive at the Earth in as little as 16 hours, but lower-energy releases may take days to make the journey.

Upon impact by a CME, the Earth’s magneto-sphere temporarily deforms, and the Earth’s magnetic field is distorted. During these disrup-tions, Earth-orbiting satellites are exposed to ionized particles, compass needles can behave erratically and electrical currents may be induced in the Earth itself. These events—geo-magnetic storms—can disrupt technical infra-structure on a global scale. Because of the risks associated with solar storms and CMEs, scien-tists constantly monitor space weather.

At a solar minimum, the estimated occur-rence of a CME is about one event every five days compared with about 3.5 per day at a solar maxi-mum. Although this may appear to put the planet in frequent jeopardy, the probability that a CME will be directed toward Earth is small. In com-parison to the Sun and the expanse of the solar

> Auroras at high latitudes. Charged particles from solar wind and geomagnetic storms follow the Earth’s magnetic field lines and can ionize gases in Earth’s upper atmosphere. Ionized oxygen molecules emit green to brownish-red light; ionized nitrogen emissions are blue or red. The aurora borealis (left) was photographed from the International Space Station over the Midwest US on January 25, 2012. The photograph of the aurora australis (right) captured by the NASA IMAGE satellite on September 11, 2005, was taken four days after a solar flare. The aurora encircles the South Pole and would appear as a curtain of light if observed from ground level. (Photographs courtesy of the NASA International Space Station and IMAGE Science Center.)

> CME image captured from space on October 22, 2011. The Large Angle and Spectrometric Coronagraph (LASCO), on board the NASA SOHO satellite, captured this image in which plasma was hurled in the direction of Mars. The Sun is obscured by a disk that allows the instrument’s sensor to focus on the emissions from the Sun’s surface, which enhances the observation of the corona by blocking direct light from the Sun. The white circle on the disk represents the size and location of the Sun’s surface. (Photograph courtesy of the SOHO EIT consortium.)

Sun’sdiameter

Autumn 2013 55

system, the Earth is tiny; most solar storms fire harmlessly away from Earth or deliver only a glancing blow.

But CMEs do strike the Earth. The Carrington event is not the only CME that has directly impacted Earth. In 1984, US President Ronald Reagan was airborne in the presidential plane Air Force One over the Pacific Ocean during a solar storm. The storm disrupted high-frequency radio communication for several hours and effec-tively isolated Air Force One from the rest of the world. In July 1989, a portion of Quebec, Canada, was blacked out for more than nine hours because a solar storm overloaded circuit break-ers on the power grid. More than 200 related events were reported across North America. The US National Academy of Sciences reported that had the storm been a Carrington-class event, cost could have ranged from US$ 1 to 2 trillion in dam-age to critical infrastructure, and recovery could have taken 4 to 10 years.16

Forecasting Space WeatherTechnologies that are sensitive to changes in the near-Earth electromagnetic environment caused by geomagnetic storms include satellite commu-nication systems, global positioning systems (GPSs), computer networks, electric grids and cell phone networks. Civilization has become increasingly dependent on these technologies, and space weather has the potential to disrupt them. Thus the need for accurate space weather forecasts has become imperative. The NOAA SWPC serves as the primary warning center for the US and provides information to the International Space Environment Service (ISES). ISES—a collaborative network of space weather providers—monitors space weather, provides forecasts and issues alerts from regional warning centers. Using a wide array of terrestrial and space-based sensors, scientists continually moni-tor the space environment for events that might impact Earth.

15. Comins NF and Kaufmann WJ: Discovering the Universe, 9th ed. New York City: W. H. Freeman and Company, 2012.

16. National Research Council of the National Academies: “Severe Space Weather Events—Understanding Societal and Economic Impacts: A Workshop Report,” Washington, DC: National Academies Press, May 2008.

17. The Lagrange points, named for Italian-French mathematician Joseph-Louis Lagrange, are the five positions where a small mass can maintain a constant pattern while orbiting a larger mass. The L1 point lies in a direct line between the Earth and Sun.

For more on the Lagrange points: “The Lagrange Points,” National Aeronautics and Space Administration, http://map.gsfc.nasa.gov/mission/observatory_l2.html (accessed August 1, 2013).

About 1.6 million km [1 million mi] from the Earth, in the general direction of the Sun, a group of NASA satellites monitors the Sun and solar wind at the L1 Lagrange point (above).17 In what is analogous to a geostationary orbit, spacecraft remain in fixed positions with the Earth’s orbit relative to the Sun. The Solar and Heliospheric

> Lagrange points. Scientists have identified five points (L1 through L5) associated with Earth’s orbit of the Sun where satellites can maintain stable orbits. These locations, called Lagrange points (green), are shown here with the gravitational potential lines (gray lines) established by the Sun-Earth system. These positions in space correspond to regions where the gravitational forces of attraction (red arrows) and repulsion (blue arrows) are in balance. The Wilkinson Microwave Anisotropy Probe (WMAP) is located around position L2, which is about 1.5 million km [930,000 mi] from the Earth. The WMAP spacecraft aligns with the Sun-Earth axis, similar to a geostationary orbit, but course corrections are required to maintain its relative position. The illustration is not to scale. (Illustration courtesy of the NASA WMAP Science Team.)

L3

L5

L1 L2

L4

Moon

Sun

Earth

56 Oilfield Review

Observatory (SOHO), the Advanced Composition Explorer (ACE) and other space-bound assets monitor the Sun’s surface and track CMEs from this position.18 Hours before a CME impact, satel-lite sentinels at the L1 point can anticipate its arrival at the Earth’s magnetosphere (left).

The SOHO satellite, launched in 1995, allows scientists to constantly monitor the Sun (below left). This satellite is one of the most reliable NASA and European Space Agency (ESA) fore-casting tools, providing scientists with data to help them forecast space weather and estimate potential consequences. The Large Angle and Spectrometric Coronagraph (LASCO), one of 12 instruments on board SOHO, records images of CMEs launched from the Sun. Using LASCO data, the SWPC has two to three days of advanced warning for the onset of geomagnetic storms.

The Solar Dynamics Observatory (SDO), devel-oped at the NASA Goddard Space Flight Center in Greenbelt, Maryland, USA, and launched on February 11, 2010, is part of a five-year NASA mis-sion to study the Sun and its influence on space weather (next page).19 Several devices are on board the satellite, including the Extreme Ultraviolet Variability Experiment and the Atmospheric Imaging Assembly. The helioseismic and magnetic imager provides real-time maps of magnetic fields on the surface of the Sun and mea-sures their strength and orientation. Changes and realignment of the Sun’s magnetic fields are early indications of potential eruptions and are crucial for the prediction of space weather and geomag-netic storms. Instruments on board the satellite can also characterize the interior of the Sun, where the magnetic fields originate. From SDO data, scientists are gaining a better understanding of solar activity and space weather.

Geomagnetic Storms BrewingGeomagnetic storms that disrupt activities on Earth are infrequent, although their conse-quences are significant; solar storms have the potential to disturb the entire planet. The tech-nologies that define modern society are suscep-tible to the effects of space weather. Induced currents can disrupt and damage modern elec-trical power grids and cripple satellites and communication systems. For the oil and gas industry, geomagnetic storms can adversely affect pipelines and supervisory control systems and disrupt surveying and geosteering opera-tions while drilling.

> The NASA Advanced Composition Explorer (ACE). Launched on August 25, 1997, the ACE satellite, a crucial component of the NASA space weather monitoring fleet, is stationed at Lagrange point L1. From this position, the satellite records radiation emitted from the Sun, the solar system and the galaxy. When bursts of solar material stream toward Earth, instruments on board ACE record the increase in the number of particles and transmit this information to scientists on Earth who use these data to warn of impending space weather events. Alerts and warnings are issued to relevant organizations and posted online by the NOAA SWPC. (Illustration courtesy of NASA.)

> Space weather monitoring by SOHO. The SOHO satellite (right) was launched in December 1995. SOHO is a joint project between the European Space Administration (ESA) and NASA to study the Sun from its deep core to the outer corona and the solar wind. The satellite weighs about 17.8 kN [2 tonUS], and its solar panels extend about 7.6 m [25 ft]. This solar eruption (left), which lasted four hours, was photographed on December 31, 2012, by the Extreme Ultraviolet Imaging Telescope (EIT) in 30.4-nm emission. Most of the plasma fell back to the Sun’s surface. The Earth is shown for scale. (Solar photograph courtesy of the SOHO EIT consortium; satellite image courtesy of Alex Lutkus.)

Relative sizeof Earth

Autumn 2013 57

The most crippling effects of geomagnetic storms come from geomagnetically induced cur-rents (GICs) that flow through electrical power grids. At the most benign level, GICs can trip cir-cuit breakers, but stronger events can destroy transformers and trigger component meltdown throughout large geographic areas.

GICs damage transformers by driving them into half-cycle saturation—the core of the trans-former is magnetically saturated on alternate half cycles. A GIC-induced voltage level of as little as 1 to 2 volts per kilometer or current of 5 amperes is sufficient to drive transformers into saturation in one second or less.20 Engineers have measured GIC currents as high as 184 amperes during geomagnetic storms; these levels are far

above that required to overload electrical grids.21 In the event of a severe GIC incident, the time required to restore damaged equipment and bring large populations back online might be measured in weeks, months or even years.

When the charged plasma cloud of a CME col-lides with Earth’s atmosphere, transient mag-netic waves alter Earth’s normally stable magnetic field; the effects can last for several days. These magnetic disturbances may cause voltage variations along the Earth’s surface, inducing electrical currents between grounding points because of the voltage potential differ-ences. GICs in this form are particularly detri-mental to transformers typically found in power plants and electrical distribution substations.

18. For more on SOHO: http://sohowww.nascom.nasa.gov/ (accessed August 13, 2013).

For more on ACE: http://www.srl.caltech.edu/ACE/ (accessed August 13, 2013).

19. For more on SDO: http://sdo.gsfc.nasa.gov/ (accessed August 13, 2013).

20. For more on detrimental effects on power grids: Barnes PR, Rizy DT, McConnell BW, Tesche FM and Taylor ER Jr: “Electric Utility Industry Experience with Geomagnetic Disturbances,” Oak Ridge, Tennessee, USA: Oak Ridge National Laboratory, ORNL-6665, September 1991.

21. Odenwald S: The 23rd Cycle: Learning to Live with a Stormy Star. New York City: Columbia University Press, 2001.

Several factors dictate the susceptibility of a given electrical power grid system to disruption and damage from solar storms. A power grid’s proximity to Earth’s polar latitudes generally increases its risk for failure or malfunction. In addition, sites located in regions of low ground

> The Solar Dynamics Observatory (SDO). The SDO satellite (left) was launched in February 2010 as part of the NASA Living with a Star Program, which studies solar variability and potential impacts on Earth and space. By examining the solar atmosphere on small scales and capturing emissions at many wavelengths simultaneously, the study hopes to determine how the Sun’s magnetic field is generated and structured and how stored magnetic energy is converted and released into the heliosphere and space. This image of the Sun’s magnetic field lines (right), captured on June 4, 2013, was taken in extreme ultraviolet light and highlights the bright coils of magnetic field lines rising up in the background above an active region. A filament, which appears as a darker region on the Sun’s surface, can also be seen. (Photograph and image courtesy of the NASA SDO.)

Filament

Magnetic field lines

58 Oilfield Review

conductivity, such as igneous rock provinces, are more susceptible to GIC effects (above).

The interconnectivity of power grids can exacerbate the potential for large-scale prob-lems. During the July 1989 solar storm, many related events were reported. These events included a transformer failure at the Salem nuclear plant in New Jersey, USA; New York Power losing 150 MW the moment the Quebec power grid went down; and the New England Power Pool, an association of power suppliers, losing 1,410 MW. Service to 96 electrical utilities

in the New England region of the US was inter-rupted before power companies could bring other reserves online.22

Damage caused by energized particles emit-ted from the Sun is not limited to terrestrial sys-tems. Satellites, space exploration vehicles and manned space missions can be affected by solar emissions, some of which are too weak to enter Earth’s magnetic field. For instance, weak solar flares and CMEs may produce solar proton events (SPEs) that are mostly unnoticed on the surface

of the Earth. However, SPEs can cause significant damage to equipment located outside Earth’s protective shield.

When high-energy charged particles collide with satellites, electrons create a dielectric charge within the spacecraft. This static charge can destroy electronic circuit boards, alter and scramble stored data and affect control instruc-tions stored in computer memory. Although these effects may result in a complete satellite failure, damage may often be corrected by simply reboot-ing onboard computers.

> Power system susceptibility. Power systems in areas with the lowest ground conductivity (left, red and darkest yellow) are the most vulnerable to the effects of intense geomagnetic activity. The high ground resistance beneath these areas facilitates the flow of geomagnetically induced currents (GICs) in power transmission lines. Auroral zones for North America are susceptible to GICs because of their proximity to polar regions. (Data from the American Geophysical Union and the Geological Survey of Canada.) For the US, scientists produced a map based on scenarios for existing power systems to determine their vulnerability to geomagnetic storms (right). Should a storm 10 times larger than the 1989 storm that disrupted power systems in Quebec arrive at Earth, the systems most at risk have been identified (red). The blue lines encircle the largest population centers served by at-risk systems. (Adapted from the National Research Council of the National Academies, reference 16.)

Typical auroralzone location

Auroral zone extremeon March 13, 1989

UNITED STATES

MEXICO

CANADA

1 to 1010–1 to 110–2 to 10–1

10–3 to 10–2

10–4 to 10–3

Regionconductivity, S/m

UNITEDSTATES

CANADA

MEXICO

Highest riskMedium riskConnected power grids

Autumn 2013 59

If the solar arrays that provide power to satel-lites are struck by high-energy protons from SPEs and CMEs, the silicon atoms in the solar cell matrix may shift positions, which increases the internal resistance of the solar cells and reduces electrical output. A single solar storm event can decrease panel life expectancy by years. If atti-tude control systems on satellites used to correct their orientation and position are damaged by high-energy particle events, a satellite can lose its orbital control, which may result in an unplanned and premature reentry into Earth’s atmosphere.23 Satellites play such a crucial role in communications that a loss could affect televi-sion, cable programming, radio service, weather data, cell phone service, automated banking ser-vices, commercial airline systems and GPS and navigation services. Routine losses as a result of satellite malfunction and premature asset failure caused by solar storms are estimated in the bil-lions of US dollars.

Consequences of solar storms may not be limited to electrical damage. The July 1989 solar storm caused compression of the Earth’s magnetosphere, reducing its typical depth of more than 54,000 km [33,500 mi] to less than 30,000 km [18,640 mi], well inside the Earth’s geosynchronous region where satellites orbit. As the Earth’s atmosphere was bombarded by energetic particles and compressed by the solar wind, the density of the upper atmosphere increased by a factor of 5 to 10. The increased drag on low-Earth orbit satellites caused orbital decay—the U.S. Air Force Space Command reported losing track of more than 1,300 orbit-ing objects that fell to lower altitudes.24 In a separate event, on March 13, 1989, NOAA reported the loss of the GOES-7 weather satel-lite. Circuit problems caused by a shower of energized particles rendered most of its systems useless. Critical solar power arrays on GOES-7 lost 50% of their efficiency. Engineers with NASA reported many other satellites experi-enced electrical failures that temporarily shut down onboard computers.25 The storm disrupted communications on the Earth and between ground controllers and orbital satellites.

Oil and gas pipeline and distribution systems are also vulnerable. In the event of a geomagnetic storm, operators may immediately lose supervi-sory control and data acquisition (SCADA) sys-tems. Operators must also consider the long-term effects associated with increased pipeline corro-sion rates. Cathodic protection systems used on pipelines to minimize corrosion maintain a nega-tive potential with respect to the ground. During solar storms, GIC events in a pipeline reduce the

effectiveness of the cathodic protection, which may increase long-term corrosive effects.26 The level of impact is affected by the specifics of pipe construction materials, pipeline diameter, bends, branches, insulated flanges and the integrity of insulation materials.

Operators are also concerned about the large percentage of modern oil and gas wells that are drilled directionally. Drillers must use strict well trajectory plans to control borehole position rela-tive to the reservoir and to avoid collision with nearby wellbores. Directional drilling relies on instruments that make real-time measurements to determine and track the subsurface location of the drilling assembly. Triaxial magnetometers measure the strength of the Earth’s magnetic field, and triaxial accelerometers are used to correct magnetometer data for position, motion and orientation. Gyrocompasses—using gyro-scopes and the rotation of the Earth to find geographic north—are also deployed on wire- line to acquire precise directional surveys.27 Disturbances in the Earth’s magnetic field aris-ing from electric currents flowing in the iono-sphere and the magnetosphere can affect these measurements (above). Mirror currents may also be induced in the Earth and oceans by variations in the Earth’s magnetic field. These external magnetic fields are affected by the solar wind,

the interplanetary magnetic field and the Earth’s magnetic core. Well placement engineers must be acutely aware of geomagnetic disturbances and variations in Earth’s magnetic field to ensure proper borehole placement.28 (See “Geomagnetic Referencing—The Real-Time Compass for Directional Drillers,” page 32.)

The Earth’s climate is also susceptible to space weather and to particle emissions from the Sun. Although the Sun appears to be a constant energy source, scientists have demonstrated

> Geomagnetic storms and directional drilling. Directional drillers use MWD tools to determine drillbit orientation and position; these measurements depend on data derived from magnetometers and accelerometers. During geomagnetic storms, magnetometers may provide erroneous readings. A solar storm occurred while an operator drilled a North Sea well, and the MWD drilling azimuth measurement (blue) was affected by the geomagnetic storm. Engineers corrected the data using a technique developed by the British Geological Survey that adjusts for space weather. The results provided a more accurate well location (green). (Adapted from Clark and Clarke, reference 28.)

259

260

261

262

263

264

Azim

uth,

deg

ree

3,600Depth, ft

Magneticstorm

4,7004,6004,5004,4004,3004,2004,1004,0003,9003,8003,700

Corrected azimuthDrilling azimuth

22. North American Electric Reliability Corporation (NERC): “Effects of Geomagnetic Disturbances on the Bulk Power Systems,” Atlanta, Georgia, USA: NERC (February 2012).

23. Odenwald, reference 21.24. Alexander, reference 14.25. Odenwald, reference 21.26. Zurich Financial Services Group: “Solar Storms:

Potential Impact on Pipelines,” http://www.zurich.com/internet/main/SiteCollectionDocuments/insight/solar-storms-impact-on-pipelines.pdf (accessed September 5, 2013).

27. Ekseth R and Weston J: “Wellbore Positions Obtained While Drilling by the Most Advanced Magnetic Surveying Methods May Be Less Accurate than Predicted,” paper IADC/SPE 128217, presented at the IADC/SPE Drilling Conference and Exhibition, New Orleans, February 2–4, 2010.

28. Clark TDG and Clarke E: “2001 Space Weather Services for the Offshore Drilling Industry,” poster presentation in Proceedings from the ESA Space Weather Workshop: Looking Towards a Future European Space Weather Program. Noordwijk, The Netherlands, December 17–19, 2001.

60 Oilfield Review

that the base energy output of the Sun varies up to 0.5% on short timescales and 0.1% over the 11-year sunspot cycle. Considered significant by atmospheric scientists, these fluctuations can affect Earth’s climate. Variations in plant growth have been correlated with the 11-year sunspot cycle and 22-year magnetic period of the Sun, as evidenced in tree ring records.29

Although the solar cycle has been relatively steady during the last 300 years, during a 70-year period in the 17th century, few sunspots were observed. This period, referred to as the Maunder Minimum, also coincided with the timing of the Little Ice Age in Europe. Some scientists have theorized that this is evidence of a Sun-Earth cli-mate connection (above).30 Recently, scientists have proposed a more direct link between the Earth’s climate and solar variability. For instance, the stratospheric winds near the Earth’s equator

change direction with each solar cycle. Studies are underway to determine how this wind rever-sal affects global circulation patterns, weather and climate.31

The Next Big EventGeomagnetic storms, although infrequent, can severely impair critical infrastructures of modern society. Because we are increasingly dependent on susceptible technologies in our intercon-nected global economy, solar storms have the potential to create havoc on a worldwide scale. The scientific community is working to improve its understanding of the technical aspects of this threat and the related vulnerabilities in various industry segments to better manage risk.

The science of space weather forecasting is still in its infancy. Scientists cannot accurately forecast the number of sunspots before the start of a solar cycle or predict geomagnetic storm activity, although some organizations do make attempts. A decade ago, before the start of Cycle 24, some forecasters were predicting the most intense solar maximum in 50 years and that the cycle might result in devastating geomag-netic storms.32 But those forecasts were wrong.

The sunspot activity of Solar Cycle 24 has been the lowest in more than 100 years, barely half the activity level of Cycle 23. Some scientists speculate that the Sun is entering another quiet period similar to the Maunder Minimum and are

asking questions: Will global climate effects be similar to those of the Little Ice Age during the Maunder Minimum or is there no direct correla-tion between sunspots and terrestrial climate? Or is this just the quiet before the storm? Even during a relatively low-amplitude solar cycle, a CME can be triggered that makes a direct hit on planet Earth.

The recurrence probability of the 1859 Carrington event is estimated at 1 in 500 years, and the recurrence probability of the 1989 Quebec storm is estimated at 1 in 150 years.33 Although scientists, engineers and risk managers are concerned about the potential damage of another Carrington-type event, they have many more tools at their disposal to help them predict and react when such an event occurs. These tools allow the scientific community to remain vigilant to the Sun’s activity and be prepared to act.

The list of solar storm consequences grows in proportion to our dependence on electromagneti-cally sensitive technology systems. The SWPC and ISES, working with many national and inter-national partners, continue to develop improved monitoring and space weather modeling capabili-ties. Advances in Earth-bound and satellite-based data acquisition systems, along with modeling and a better understating of our interlinked rela-tionship with the Sun, hold promise of reducing our exposure risk when the Earth is directly in the path of the next great solar storm. —TS

> Sunspot cycles and terrestrial weather. Scientists have not reached consensus regarding the effects of solar activity on the Earth’s climate and weather. Most, however, would agree that the Sun is the primary heat source for the Earth, thus the major driver of climate. Some scientists have tried to draw a correlation between the absence of sunspots during the Maunder Minimum (top)—a 70-year period in the 17th century—and the Little Ice Age that affected much of the Earth, especially Europe (bottom). The Dalton Minimum, another period of low sunspot occurrences around 1800, corresponded to lower than average global temperatures, as well. The rise in total average number of sunspots (black) beginning in the 1900s appears to correspond to increases in global temperatures. Although a close examination of the data points to other factors producing temperature variations, such as volcanic eruptions and changes in CO2 levels, some observers propose solar activity as a major component in climate and temperature fluctuations. The activity of Solar Cycle 24 is comparable to that in the cycles around 1800 rather than those of the 20th century. A century from now, scientists may be able to look back and debunk or validate the causal relationship of sunspots to climate change.

Medieval warm period

Mean temperature

Little Ice Age

Tem

pera

ture

Date

Northern Hemisphere Temperatures over the Last 1,000 Years

900 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900 2000

20000

50

100

150

200

250

Suns

pot n

umbe

rs

Date

400 Years of Sunspot Observed Data

1600 1650 1700 1750

MaunderMinimum

DaltonMinimum

Modernmaximum

1800 1850 1900 1950

Less-reliable observation dataReliable observation data

29. For recent research on solar cycles’ effects on Earth’s weather: Meehl GA, Arblaster JM, Matthes K, Sassi F and van Loon H: “Amplifying the Pacific Climate System Response to a Small 11-Year Solar Cycle Forcing,” Science 325, no. 5944 (August 2009): 1114−1118.

30. Weng H: “Impacts of Multi-Scale Solar Activity on Climate. Part I: Atmospheric Circulation Patterns and Climate Extremes,” Advances in Atmospheric Sciences 29, no. 4 (July 2012): 867−886.

31. Weng H: “Impacts of Multi-Scale Solar Activity on Climate. Part II: Dominant Timescales in Decadal-Centennial Climate Variability,” Advances in Atmospheric Sciences 29, no. 4 (July 2012): 887−908.

32. “Solar Storm Warning,” NASA (March 15, 2006), http://www.nasa.gov/vision/universe/solarsystem/ 10mar_stormwarning.html (accessed August 18, 2013).

33. Zurich Financial Services Group, reference 26.

Contributors

Autumn 2013 61

Dalia Abdallah is a Senior Production Chemist for Abu Dhabi Company for Onshore Oil Operations (ADCO) in Abu Dhabi, UAE. She joined the company in 2008 and has focused on mitigation strategies for scales and asphaltenes, corrosion issues, produced water treat-ment and effective stimulation strategies. Previously, she worked as a Schlumberger fluid analysis engineer in Abu Dhabi. Dalia, who holds two patents and is the author of several articles, has a PhD degree in chemistry from Queen’s University, Kingston, Ontario, Canada.

Khaled Al-Hendi is a Drilling and Workover Supervisor for Kuwait Oil Company (KOC) in Ahmadi, Kuwait. He joined KOC in 2006 as a company representative overseeing the drilling and workover of wells affected by the Iraqi invasion of Kuwait. Khaled holds a degree in petroleum engineering from Kuwait University, Kuwait City.

Adel Abdulla Al-Khalaf is a Senior Petroleum Engineer and Petrophysicist with Qatar Petroleum in Doha, Qatar, working on well integrity in offshore fields. He previously worked for the company as assis-tant petroleum engineer and then as reservoir engi-neer in the Dukhan onshore field. Adel earned a BS degree in petroleum and natural gas engineering from West Virginia University, Morgantown, USA.

Zaid Al-Kindi, who is a Well Integrity Domain Champion for Schlumberger in Abu Dhabi, UAE, provides support and training for pipe integrity and zonal isolation projects in Egypt, Oman, Pakistan and the UAE. With the company since 1994, he has held positions in service quality and account management. Previously, he worked in technical sales for Galadari Heavy Equipment Company and as a project engineer in the UK. Zaid received a BS degree in mechanical engineering from King’s College London.

Abdulmohsen S. Al-Kuait is a General Supervisor of the Saudi Aramco Safaniya production engineering division in Dhahran, Saudi Arabia. During more than 25 years at Saudi Aramco, he has worked on numerous projects with a focus on production engineering and producing operations. Abdulmohsen obtained a BS degree from King Fahd University of Petroleum and Minerals, Dhahran, Saudi Arabia.

Mohannad Al-Muhailan is Team Leader Deep Drilling Engineering for Kuwait Oil Company in Ahmadi, Kuwait. He has 15 years of experience in conventional drilling and high-pressure, high-temperature drilling, workover operations and drilling management and finances. Mohannad has also worked in underbal-anced, managed pressure and cluster drilling opera-tions. He has a degree in petroleum engineering from Kuwait University in Kuwait City.

Hassan B. Al-Qahtani is a Supervisor of the Saudi Aramco Safaniya production engineering division in Dhahran, Saudi Arabia. In more than 17 years as a petroleum engineer for Saudi Aramco, he has worked on reservoir management best practices, production engineering and field development. Hassan holds a BS degree from King Fahd University of Petroleum and Minerals, Dhahran, and an MS degree from The University of Texas at Austin, USA. He is a gradu-ate of the Saudi Aramco Petroleum Engineering Technologist Development Program with a specialty in improved oil recovery.

Karam S. Al-Yateem, a Group Leader for the Saudi Aramco Transformative Technology Think Tank in Dhahran, Saudi Arabia, began with the company in 2005 as a reservoir, testing and production engineer in onshore and offshore field locations. He is a member of the SPE International Production and Operation com-mittee, has authored or coauthored several technical papers and is an SPE Certified Petroleum Engineer. Karam earned a BS degree in petroleum engineering from King Fahd University of Petroleum and Minerals, Dhahran, Saudi Arabia, and an MS degree with a specialty in smart oilfield technologies and management from the University of Southern California, Los Angeles, USA.

Anatoly Arsentiev is the Schlumberger Drilling & Measurements Electronic Team Leader and GeoMarket* Direction and Inclination (D&I) Calibration Champion in Irkutsk, Russia, respon-sible for preventive maintenance and repair of MWD and LWD tools and addressing D&I surveying tech-nicalities. He started his career with Schlumberger in 2006 as an electronics technician. Prior to that, he worked with Insight LLC, an engineering medi-cal company in Irkutsk. Anatoly holds a degree in terrestrial physics from Irkutsk State University and did postgraduate research at the Institute of Solar-Terrestrial Physics, Irkutsk.

Khalid S. Asiri is a Gas Production Engineering Supervisor for Saudi Aramco in Udhailiyah, Saudi Arabia; he is responsible for the unconventional gas production engineering unit and all unconventional stimulation activities in tight gas reservoirs. He has also worked in several areas in the company, includ-ing gas production engineering, gas well completion and services, reservoir engineering and gas drilling engineering. Before joining Saudi Aramco in 2002, he worked with the Saudi Arabia Ministry of Petroleum and Mineral Resources. Khalid received a BS degree in petroleum engineering from King Saud University, Riyadh, Saudi Arabia. He is a member of the SPE, the SPE Saudi Arabia Section and Saudi Council of Engineers.

Nausha Asrar is a Senior Materials Scientist and Manager of Materials Support and Failure Analysis for Schlumberger in Sugar Land, Texas. He currently specializes in failure analysis of downhole tools and materials testing and evaluation. Before joining Schlumberger in 2005, he was a materials and cor-rosion specialist with Shell Global Solutions. He also worked for Saudi Basic Industries Corporation and Saline Water Conversion Corporation in Saudi Arabia and for the Steel Authority of India. Nausha obtained an MS degree in chemistry from Aligarh Muslim University, Uttar Pradesh, India, and a PhD degree from Lomonosov Moscow State University.

Mohammed A. Atwi is General Supervisor for the Saudi Aramco Engineering Division at South Ghawar Production in Udhailiyah, Saudi Arabia. During his 10-year career with Saudi Aramco, he has worked in gas production engineering, well completion opera-tions, deep gas drilling engineering and reservoir man-agement. Mohammed has a BS degree in petroleum engineering from The University of Tulsa.

Syed Aamir Aziz is a Senior Production Engineer with Schlumberger in Dhahran, Saudi Arabia, where he conducts the processing and interpretation of production logs and well integrity monitoring. He began his career in 2002 as a wireline field engineer with National Petroleum Services in Saudi Arabia, where he later became a log analyst and geoscientist responsible for log processing and interpretation. He joined Schlumberger in 2008 in Abu Dhabi, UAE. Aamir received BS and MS degrees in geology from the University of Karachi, Pakistan.

Abderrahmane Benslimani is the Associate Well Integrity Domain Champion for Schlumberger Wireline in Ahmadi, Kuwait. He began his career in 2004 in the UAE as a logging field engineer and held field opera-tions positions in Algeria, Libya and China before moving to Kuwait in 2012. Abderrahmane holds a diploma in electrical engineering from the National Polytechnic School of Algiers, Algeria, MS degrees in electrical sciences and electrical engineering, both from Université de Montpellier II, France, and an MS degree in mathematics and computer science from Université Paris-Sud, France.

Andrew Buchanan is the Senior Operations Geologist with Eni US Operating Company Inc. in Anchorage, where he has been since 2009. He previ-ously worked for ASRC Energy Services as a geology consultant. Andrew earned a BS degree in geol-ogy from the University of Regina, Saskatchewan, Canada. He currently serves as Past President of the Petroleum Club of Anchorage.

Mohamed Fahim is a Petroleum Engineering Expert for the Abu Dhabi Company for Onshore Oil Operations Technical Services division in Abu Dhabi, UAE. Previously, Mohamed worked as a senior petroleum engineer for the Gulf of Suez Petroleum Company in Egypt. He holds a BS degree in petroleum engineering and is an SPE Certified Petroleum Engineer.

Carol A. Finn is the Geomagnetism Group Leader for the US Geological Survey (USGS) Geologic Hazards Science Center in Denver, where she has worked since 2006. She is responsible for operations and mainte-nance of 13 USGS geomagnetic observatories in the US and its territories. Before joining the USGS, she served 10 years with the US Air Force Technical Applications Center as a research geophysicist and was a geodesist with the US Defense Mapping Agency Hydrographic/Topographic Center. Carol received an MS degree in geophysics from St. Louis University, Missouri, USA, and a BS degree in geology from Southwest Missouri State University, Springfield.

M. Aiman Fituri is the Schlumberger Wireline Well Integrity Domain Champion in Doha, Qatar. Before his current position, he supported petrophysical openhole logging and cementing evaluation. He joined the com-pany in 2000 in Oman and held operations positions in India, Sudan and Egypt before moving to Qatar. Aiman obtained a BS degree in computer engineering from Al-Fateh University, Tripoli, Libya.

David H. Hathaway is an Astrophysicist who served as the head of the solar physics group at the National Aeronautics and Space Administration (NASA) Marshall Space Flight Center, Huntsville, Alabama, USA, from 1996 to 2010. He was a postdoctoral fellow at the National Center for Atmospheric Research in

62 Oilfield Review

Boulder, Colorado, USA, and a research associate and an assistant astronomer at the National Solar Observatory, Sunspot, New Mexico, USA, before coming to Huntsville in 1984. He holds a BSc degree in astron-omy from the University of Massachusetts in Amherst, USA, and an MSc degree in physics and a PhD degree in astrophysics from the University of Colorado, Boulder. David has written more than 150 papers and has three US patents, including two for the NASA Invention of the Year in 2002.

Ram Jawale is a Drilling Engineer for Kuwait Oil Company in Ahmadi, Kuwait. He began his career with Gujarat State Petroleum Corporation in Gandhinagar, Gujarat, India, as a drilling engineer for the Kingston Group Offshore high-pressure, high-temperature proj-ect. In that capacity, he performed engineering plan-ning and execution for exploratory drilling. Ram has a degree in petroleum engineering from Maharashtra Institute of Technology, Pune, India.

Oscar Jiménez Bueno, who is based in Villahermosa, Mexico, joined Petróleos Mexicanos (PEMEX) in 1984 as a reservoir engineer. Within the company, he has held multiple engineering positions, working on asset develop-ment, reservoir stimulation and production optimiza-tion. He has been responsible for well completions with production of more than 3,180 m3/d [20,000 bbl/d]. Oscar obtained a BS degree in petroleum engineering and an MSc degree in reservoir engineering from Universidad Nacional Autónoma de México in Mexico City.

J.J. Kohring has been the Well Integrity Domain Champion for Schlumberger in Dhahran, Saudi Arabia, since 2010. He began his career with Schlumberger Wireline in 1979 and for the past 15 years has special-ized in borehole production and well integrity. Jim is a Principal Production Engineer with experience in the Middle East, the US, Nigeria and Indonesia. He holds a BS degree from the University of Alaska, Fairbanks, USA.

Fraser Lawson is a Well Engineering Supervisor for Tullow Ghana Ltd. in Accra, Ghana. He has 29 years of well engineering experience, including extended-reach drilling and deepwater projects. Fraser earned a BSc degree (Hons) in civil engineering from Heriot-Watt University, Edinburgh, Scotland, and an MSc degree in petroleum engineering from the University of Strathclyde, Glasgow, Scotland, and is a Chartered Engineer.

Bruno Lecerf, based in Sugar Land, Texas, is a Product Engineering Manager with the Pressure Pumping and Chemistry group within Schlumberger Engineering, Manufacturing and Sustaining. Previously, he was a project manager at the Novosibirsk Technology Center, Russia, and prior to that, a solutions engineer for acidizing at the Integrated Productivity and Conveyance Center in Sugar Land. Bruno obtained an MS degree in chemistry from Ecole Supérieure de Chimie Physique Electronique de Lyon, France, and an MS degree in chemical engineering from the University of Houston.

Tim Lesko is a District Technical Engineer with the Schlumberger Disruptive Advances for Sustainable Unconventional Stimulation project in Conway, Arkansas, USA. He began his Schlumberger career in 2004 as a product development engineer with the Pressure Pumping and Chemistry group in Sugar Land, Texas. In 2006, he transferred to the Novosibirsk Technology Center in Russia, where he worked with the MaxCO3 Acid* team. He has worked on projects such as fiber diversion in shales, proppant conductivity

and fracture stimulation water quality. Tim received a BS degree in chemical physics from the University of California, Riverside, and a PhD degree in chemistry from the California Institute of Technology, Pasadena.

Rodney W. Lessard, who joined Schlumberger in 2001 after completing postdoctoral work at Purdue University, West Lafayette, Indiana, USA, is Senior Production Simulation and Optimization Scientist in Houston. He has coauthored several papers on very high-energy gamma rays, astronomy and oilfield portfolio management. Rod has BSc and MSc degrees in physics from the University of Calgary and a PhD degree in experimental physics from the National University of Ireland, Dublin.

Jeffrey J. Love, who joined the USGS in 2001, is a Research Geophysicist and USGS Advisor for Geomagnetic Research in Denver. He also teaches geophysics at the Colorado School of Mines, Golden. He has held research positions at the University of Leeds, England; Scripps Institution of Oceanography, La Jolla, California; and Institut de Physique du Globe de Paris. Jeffrey earned a BA degree in physics and applied mathematics from the University of California, Berkeley, and a PhD degree in geophysics from Harvard University, Cambridge, Massachusetts.

Stefan Maus is the Director of Magnetic Variation Services LLC and is a Senior Scientist at the University of Colorado, Boulder; he also maintains a laboratory at the US National Geophysical Data Center in Boulder. Previously, he was a scientist at GFZ Potsdam, Germany, and a lecturer at the Free University, Berlin. Stefan holds a BSc degree in mathematics and an MSc degree in geophysics, both from Ludwig Maximilian University of Munich, Germany, and a PhD degree in geophysics from Osmania University, Hyderabad, Andhra Pradesh, India.

Fred Mueller is the North America Engineering Director for Network of Excellence in Training (NExT, a Schlumberger company) in College Station, Texas. In 1980, he joined Dowell, which is now part of Schlumberger, as a field engineer. He spent many years with Schlumberger technical support systems for pro-duction enhancement and optimization and was a Well Services technical manager in California and South Texas. Fred has experience in the technical and opera-tional aspects of cementing and hydraulic fracture design for tight gas and shale formations. He received a BS degree in engineering technology from Texas A&M University, College Station.

Shola Okewunmi worked as a directional drilling subject matter expert with Chevron Energy Technology Company in Houston and now works in Jakarta as a Senior Drilling and Completions Engineer for deepwa-ter development projects. He has more than 20 years of experience in drilling and measurements, geosteering and formation evaluation. Shola obtained a BS degree in mechanical engineering from Obafemi Awolowo University, Ile-Ife, Nigeria, an MBA degree from the University of Houston-Victoria and a PhD degree in engineering management from Kennedy Western University, Wyoming, USA.

Alejandro Peña is a Global Chemistry and Materials Portfolio Manager for Schlumberger in Sugar Land, Texas. He oversees the corporate strategy for innova-tion in chemistry-enabled well stimulation technolo-gies. He earned his BS degree in chemical engineering and was an assistant professor at Universidad de Los

Andes in Mérida, Venezuela. After completing his PhD degree in chemical engineering at Rice University in Houston, he joined Schlumberger as a senior chemical engineer. Since then, he has held several operational, engineering and technology management positions within Schlumberger in North and South America. Alejandro holds several patents and has authored various publications on interfacial phenomena and reservoir stimulation technology.

Alexandre Z. I. Pereira is a Petrobras Advisor in the Well Engineering Group in Rio de Janeiro and is a Specialist in chemical treatments. He worked in the Campos Basin Operational Unit when he joined Petrobras in 1987 and then moved to the Rio de Janeiro Operational Unit, where he develops projects in completion, corrosion, scaling and stimulation at the Petrobras Research and Development Center. A member of the SPE, Alexandre holds a BS degree in chemical engineering and an MA degree in chemistry from the Universidade do Estado do Rio de Janeiro.

Benny Poedjono is the Schlumberger Surveying and Risk Management Manager for North America Offshore and a Petro-Technical Engineer Advisor for Wellbore Positioning, in Sugar Land, Texas. He started his career with the company in 1982 as a Wireline field engineer and has held operations, management, technical support and business devel-opment positions in 22 countries. In the past 10 years, he has been concentrating on advanced well surveying and collision avoidance management. He holds several patents and trade secrets and has pub-lished more than 30 technical papers. Benny has a BS degree in electronics engineering from the Bandung Institute of Technology, Indonesia.

Mahmut Sengul is a Schlumberger Production Technology Advisor in Houston. He joined the company in 1997 as a reservoir solutions manager in the UAE, where he was involved in enhanced oil recovery project design. He then became vice president of Schlumberger Carbon Services in the Middle East. Prior to his Schlumberger career, Mahmut worked for the Turkish Petroleum Corporation, Mobil and ADCO. Mahmut received a BS degree in petroleum engineering and an MS degree in reservoir engineering, both from the Middle East Technical University, Ankara, Turkey.

Fernanda Tellez Cisneros is a Schlumberger Senior Design Engineer for matrix acidizing, acid fracturing and hydraulic fracturing in Villahermosa, Mexico. She began her career as a Well Services field engineer in 2007. Fernanda earned a BS degree in chemical engineering from Instituto Tecnológico y de Estudios Superiores de Monterrey, Mexico.

E. William Worthington is a Geophysicist in Geomagnetic Observatory Operations with the USGS Geologic Hazards Science Center in Golden, Colorado. He has worked for the USGS since 1988. During his career, he has been a visiting scientist at the Soviet Academy of Sciences, a researcher with the Geophysical Institute at the University of Alaska, Fairbanks, and chief of the USGS College Magnetic Observatory in Fairbanks. Bill received his BS degree in geosciences from the University of Arizona, Tucson, USA, and MS and PhD degrees in geophysics from the Colorado School of Mines, Golden.

An asterisk (*) is used to denote a mark of Schlumberger.

The term production logging covers a wide array of sensors, measurementsand interpretation techniques. Operators use production logs to evaluate fluid movement in and out of wellbores, quantify flow rates and determine fluid properties at downhole conditions. Completion engineers can evaluate production and perforation efficiency and plan remediation or modify future completion designs based on the interpretation of production logs. Reservoir and production engineers can use these logs to help manage and optimize well and reservoir performance.

Production logging traces its origin to the 1930s and the measurement of wellbore temperature. Over the decades, other measurements—including pressure, fluid density, flow velocity and holdup (volume fraction of pipe occupied by fluid)—have been added to the service. Although measure-ments of pressure, temperature and flow rate can be obtained at the sur-face, surface measurements do not necessarily reflect what is happening in the reservoir. Comprehensive production log evaluation requires that mea-surements be acquired downhole.

Production Logging Measurements Production logging consists of several measurements, many of which are used in a complementary capacity to determine fluid and flow properties (below). Fluid velocity is commonly measured with a spinner flowmeter—a rotating blade that turns when fluid moves past it. In ideal conditions, the rotational speed of the blade in revolutions per second (RPS) is propor-tional to the fluid velocity. Friction in the spinner bearings and effects from fluid viscosity result in nonlinear velocity responses, requiring calibration of the measurement. This calibration is accomplished by making upward and downward passes at varying logging speeds. Before absolute fluid velocity is computed, spinner speed is corrected for relative tool speed. Because of friction near the pipe wall, absolute fluid velocity is not the same as the average velocity of fluid moving through the pipe. After applying correction factors, engineers convert the spinner velocity to an average velocity using

computer modeling techniques, which present the fluid velocity profile across the pipe diameter.

Pressure is a versatile measurement with several applications for reser-voir and production engineers. Strain, sapphire and quartz gauges are the main devices used to measure pressure. Engineers may also measure pres-sure using a manometer—a device that converts mechanical displacement to pressure. From wellbore pressure data, engineers can determine well stability at the time of logging, estimate reservoir pressure from multirate logging surveys, calculate fluid density and obtain key reservoir parameters by performing transient rate analyses.

Temperature is an integral measurement for all production logging. Engineers use temperature data to make qualitative conclusions about fluid entries, particularly in low–flow rate scenarios in which a spinner may not be sensitive enough to detect movement. Gas entries create cooling anoma-lies that are easily detected using temperature logs. Temperature measure-ments are also used in fracture treatment evaluation and to evaluate injection well performance. Using temperature data, engineers may be able to evaluate the integrity of well completions, detect casing leaks and iden-tify flow through channels behind pipe. Resistance temperature detectors, the most common type of sensor, usually consist of a platinum wire or film deposited on a nonconductive surface. Changes in temperature cause varia-tions in resistance, which is calibrated and converted to temperature.

Fluid density measurements differentiate oil, gas and water. Service companies have developed tools based on a variety of physical principles to obtain fluid density measurements:• differential pressure across two ports separated by a known distance• Compton scattering of gamma rays• pressure gradient relation to density• flow vibration relation to density and viscosity.

In the case of two-phase flow, engineers can use fluid density—in con-junction with other measurements such as fluid viscosity—to compute holdup. Where multiphase flow is present, they must employ tools with probes distributed across a wellbore to directly measure the fluid holdup. One type of tool senses differences in optical reflectance to obtain holdup, which involves measuring the amount of light reflected back from a fluid. Another type of tool differentiates water from oil and gas using probes that measure electrical properties of the fluids.

Auxiliary measurements commonly acquired by production logging strings are casing collar logs, gamma ray logs, caliper and deviation. Casing collar and gamma ray logs provide depth control and correlation with com-pletion components. Caliper and deviation data are used in production mod-eling programs.

Production logs can be difficult to interpret because fluid flow may not be uniform, and multiple passes result in large amounts of data, some of which may produce conflicting answers. Computer programs have been developed to assist engineers in understanding downhole conditions; computer-generated interpretations remove some of the

Autumn 2013 63

DEFINING PRODUCTION LOGGING

Principles of Production Logging

Oilfield Review Autumn 2013: 25, no. 3.

Copyright © 2013 Schlumberger.

Parijat MukerjiProduction Logging Advisor

> Production logging toolstring. This production logging toolstring consists of a fullbore spinner, fluid holdup and bubble count probes, a pipe diameter caliper and centralizer, a relative bearing sensor, pressure gauges, a temperature sensor, a gamma ray tool and a casing collar locator. When engineers run the tool in memory mode, batteries and a data recorder are used. Surface readout tools use a telemetry and power section.

Basic Measurement Sonde Spinner FlowmeterBatteries, recorder, casing collarlocator and sensors to measure

gamma ray, temperature and pressure

Caliper, water holdup,bubble count,

relative bearing, centralizer

Oilfield Review64

DEFINING PRODUCTION LOGGING

ambiguities associated with the interpretation process (below). The interpretation product can often help the engineer identify more-pro-ductive intervals, detect water entry and determine intervals that do not contribute to production.

Flow RegimesTo analyze production logging data, production engineers must be aware of downhole flow regimes. Knowledge of expected flow regimes allows opera-tors to choose measurements suitable for the downhole conditions.

Single-phase flow—when only oil, gas or water is produced—is the sim-plest flow to evaluate; however, it is uncommon in most wells. Two- and three-phase flow—mixtures of two or three fluid types—can exhibit a wide variety of flow regimes and are complicated by deviated wellbores (right). In such cases, gravity ensures that the lighter phase travels at a higher velocity than the denser phases. The velocity difference between the differ-ent fluids is the slip velocity.

As fluids migrate uphole, the flow regime usually changes. For instance, oil with dissolved gas can enter the wellbore as a single phase. The hydro-static pressure decreases as the oil rises, and gas bubbles begin to form in the oil. The flow regime is then bubble flow. Pressure is further reduced as the mixture moves uphole; more bubbles form and smaller bubbles aggre-gate to create larger bubbles. Large bubbles, or gas slugs, travel faster than both small bubbles and liquids. Gas slugs may initiate slug flow. Slugs tend to unite and move to the center of the pipe, forcing most of the oil to flow along the pipe walls. This flow regime is called froth flow. Eventually, a mist flow regime may be reached, in which the gas is carrying droplets of

oil, and both fluids are traveling at essentially the same velocity. This simpli-fied example becomes complex in deviated wellbores or when fluids of vari-ous phases enter the wellbore from multiple zones. Modeling programs attempt to resolve these complexities using production log data.

Reservoir Surveillance and ManagementProduction logs help engineers diagnose problems that occur during the life of a single well and are also used for management and surveillance of mul-tiple wells or of the entire reservoir. A common challenge faced by operators in mature fields is identifying zones that produce excessive amounts of water. Produced water impacts surface handling operations because water must be separated from the flow stream for disposal. Produced water may also affect well performance by limiting the volume of hydrocarbons being produced. After identifying water-producing zones, production engineers may perform remediation to reduce or eliminate water production.

Operators may also use production logs to calibrate reservoir simulation models. During full-field reservoir model history-matching, engineers can use production log data to help identify or match zonal contributions, fine-tune parameters and align the model with the empirical performance data.

Lifetime LoggingProduction logs provide reservoir and production engineers with a diag-nostic aid for understanding the downhole wellbore environment. These in situ measurements acquired under dynamic conditions are a snapshot of the existing situation. But that snapshot captures the situation only for that moment in time. Whether by fluid extraction or injection, oil and gas production changes reservoir conditions. Production logs help operators understand well and reservoir dynamics over the life of a well and create a roadmap for future remediation, production enhancements and reservoir development programs.

> Fluid flow. Theoretical work and flow loop experiments have helped engineers understand multiphase flow in vertical, deviated and horizontal wellbores. The parameters of interest include pipe diameter and inclination and fluid density, viscosity and flow rate. Each case shows the variation in water and oil holdup based on well deviation.

Near-Vertical Well Near-Horizontal WellDeviated Well

• Oil at the top, water at the bottom and a mixture of the two in the middle of the pipe

• Almost stratified flow structures

• Water phase at the bottom of the pipe• Dispersed oil phase in the uppermost level of the pipe

• Highly complex flow structures

• Oil and water mixed across the section of the pipe

> Production log. This log shows data typically provided by downhole sensors in a production logging survey. Two intervals have open perforations (Track 1, red). Engineers make multiple passes at different logging speeds (Track 2); negative cable speeds represent down passes and positive logging speeds are up passes. Color-coding based on logging speed helps differentiate datasets. Gamma ray data (Track 3) provide correlation with openhole logs. From spinner data (Track 4), engineers identify changes in fluid velocity associated with inflow from production, outflow from thief zones or outflow from injection. The fluid density data (Track 5) indicate water (1.0 g/cm3) below the perforations (the sump), which then transitions to mainly oil (0.75 g/cm3). Temperature data (Track 6) indicate heating or cooling effects caused by inflow of fluids. Steady well pressure (Track 7) is a characteristic of stable flow during acquisition. Holdup data (Track 8) indicate water and oil fraction within the wellbore. The software computes incremental and cumulative flow rates (Tracks 9 and 10). The two intervals are producing oil; the lower interval is also producing a trace amount of water (Track 9).

GammaRay

CableSpeed

X,900

60 ft/min

90 ft/min

1,200 ft/min

–120 120 0 0 1.1g/cm3 194 196 0.8

Oil Water Oil Water

1.03,600 0 3,000bbl/d 0 5,000bbl/d3,710psi°F–15 350RPS250gAPI

Y,000

Perfo

ratio

ns

Dept

h, ft

CumulativeProductionProduction

WaterHoldup

WellPressure

FluidTemperature

FluidDensitySpinner